Todos los marcadores

¿Está listo para empezar a usar la IA?  Microsoft Reactor proporciona eventos, formación y recursos de la comunidad para ayudar a startups, emprendedores y desarrolladores a crear su próximo negocio con tecnología basada en IA. ¡Acompáñenos!

¿Está listo para empezar a usar la IA?  Microsoft Reactor proporciona eventos, formación y recursos de la comunidad para ayudar a startups, emprendedores y desarrolladores a crear su próximo negocio con tecnología basada en IA. ¡Acompáñenos!

Te mostramos marcas peruanas bravazas, guías de marcas de acuerdo a diferentes temáticas y lxs diseñadoxs nacionales que contribuirán positivamente a la moda peruana

Advances in computing power, natural language processing, and digitization of text now make it possible to study a culture’s evolution through its texts using a ‘big data’ lens. Our ability to communicate relies in part upon a shared emotional experience, with stories often following distinct emotional trajectories and forming patterns that are meaningful to us. Here, by classifying the emotional arcs for a filtered subset of 1,327 stories from Project Gutenberg’s fiction collection, we find a set of six core emotional arcs which form the essential building blocks of complex emotional trajectories. We strengthen our findings by separately applying matrix decomposition, supervised learning, and unsupervised learning. For each of these six core emotional arcs, we examine the closest characteristic stories in publication today and find that particular emotional arcs enjoy greater success, as measured by downloads.

Bring your own code, and run CPU, GPU, and data-intensive compute at scale. The serverless platform for AI and data teams.

In 1984, Tom Zito interviewed a 29-year-old Steve Jobs, months after he introduced his first Mac.

A router for LLMs and other AI models

Create diagrams, design docs, and visual documentation with your team. Your all-in-one technical ideation tool.

Overview of automated tools for catching: low-quality responses, incomplete/vague prompts, and other problematic text (toxic language, PII, informal writing, bad grammar/spelling) lurking in a instruction-response dataset. Here we reveal findings for the Dolly dataset.

🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming - geekan/MetaGPT

Dingboard, the innovative meme making and AI-powered image editing platform. Create, edit, and share memes effortlessly.

Use our Online Video Trimmer to cut a fragment from any video! It supports any video format - MP4, AVI, 3GP, and many more! You can clip video files up to 4 GB! Try our free tool today!

Diagram as Code

Integrate cutting-edge LLM technology quickly and easily into your apps - microsoft/semantic-kernel

Genere e implemente sitios híbridos de Next.js con Azure Static Web Apps.

A game where you need to pick the design that is most correct. Test your attention to details!

The Speech service supports numerous languages for speech to text and text to speech conversion, along with speech translation. This article provides a comprehensive list of language support by service feature.

An Astra DB and OpenAI chatbot. Contribute to datastax/ragbot-starter development by creating an account on GitHub.

Starter project for creating a chatbot using Astra DB and OpenAI.

Clone a voice in 5 seconds to generate arbitrary speech in real-time - CorentinJ/Real-Time-Voice-Cloning

Everything you need to take your system design skill to the next level

Stop grappling with data, scouring for the perfect React library, and coding everything from scratch. Build internal tools 10X faster with Appsmith.

Instantly turn your Database into a No-Code Platform

Grist is a relational spreadsheet-database that empowers you to organize your data, your way. Get started with a free trial.

Self-hosted open source no-code online database and Airtable alternative. Baserow gives you the powers of a developer without leaving your browser.

A crisp set of 15×15 icons designed by the WorkOS team.

Build React-based internal tools, admin panels, dashboards & B2B apps with unmatched flexibility.

La Dirección Desconcentrada de Cultura de Cusco presenta el libro titulado «Machupicchu Investigaciones Interdisciplinarias» que consta de dos volúmenes, en los cuales se presenta información resultante de las investigaciones científicas que se ejecutan de manera permanente en el Parque Arqueológico Nacional de Machupicchu, incluyendo artículos publicados e inéditos, con mayor incidencia respecto a la llaqta inka.

In corporate strategy projects, executive leadership teams work through a series of questions to determine how their businesses can succeed. Individuals can use a similar process to figure out how to live a meaningful life. It starts with defining what makes a great life for you and then outlining your purpose and vision. You must also look at your current “portfolio” — the areas in which you spend your time and energy — to see if you’re investing the best of yourself in the activities most important to you. You should consider what research says about how people tend to find meaning and joy in life. Finally, you’ll want to identify areas where you need to make changes, and then ensure you follow through with objectives and key results. This program, Strategize Your Life, has been tested with more than 500 people around the world. With a few hours of work, you can develop a personal life strategy and summarize it on a single page.

Create more effective presentations, faster. Tome's AI, personalization tools, and analytics make it easy to create sales and marketing materials that stand out.

Decktopus is an AI presentation maker, that will create amazing presentations in seconds. You only need to type the presentation title and your presentation is ready.

Create diagrams and plan your code with TypeScript.

Dark Mode Design is a showcase of beautifully designed and inspiring dark mode websites. Dim the lights, lower your screen brightness, and enjoy.

Awwwards are the Website Awards that recognize and promote the talent and effort of the best developers, designers and web agencies in the world.

Camille Mormal is an Independent Interactive Designer & Art Director – Specializing in building art direction for immersive & interactive web experiences.

A 5kB WebGL globe library.

Getting Query Assistant from concept to feature diverted R&D & marketing resources. So, did investing in LLMs do what we wanted it to do?

Build gen AI models with Together AI. Benefit from the fastest and most cost-efficient tools and infra. Collaborate with our expert AI team that’s dedicated to your success.

Last week, I did a Q&A session for a friend's security class. One of the students asked a question that I loved. They asked something like, "As a principal engineer, how do you make sure your company stays at the forefront of innovation?"

Join 8000+ readers to master AI Engineering and stay ahead of the curve! 🚀

When a group of libertarians set about scrapping their local government, chaos descended. And then the bears moved in.

To extract main article from given URL with Node.js - extractus/article-extractor

Top performance in a tiny package. Free, open-source and feature-rich. At just 45 kilobytes, the dream of super compact interactive charts is now a reality.

Suppose you were copied into a non-biological substrate, and felt as intelligent and as conscious as you currently feel now. All questions of identity aside, do you think this new version of you has moral weight? We do.

The JavaScript library for bespoke data visualization

Three weeks ago we launched OpenGPTs, an implementation of OpenAI GPTs and Assistant API but in an open source manner. OpenGPTs allows for implementation of conversational agents - a flexible and futuristic cognitive architecture. One large part of agents is memory. In their current implementation, GPTs, OpenGPTs, and the Assistants

Nendo is the professional AI Audio Platform that unleashes the power of sonic intelligence & generation.

Run fast, low-latency inference tasks on pre-trained machine learning models natively on Cloudflare Workers.

The most advanced AI upscaler & enhancer. Magnific can hallucinate and reimagine as many details as you wish guided by your own prompt and parameters!

Vmake is an online image & video editing studio that makes creating product photos and social media content easier than ever.

Deploy backend applications and databases in minutes

We are building a future where anyone can make great music. No instrument needed, just imagination. From your mind to music.

Uploadcare provides companies with simple, powerful, developer-friendly building blocks to handle file uploading, processing, and delivery. A complete out-of-the-box solution, built for engineers by engineers.

Simple, beautiful database design tool for developers to create, collaborate and visualize their entity relationship diagrams.

Nature - The meta-learning for compositionality approach achieves the systematicity and flexibility needed for human-like generalization.

A compilation of evidence that Hal Finney and Satoshi Nakamoto were different people.

Capture reality with Polycam's LiDAR scanner & photogrammetry. Make 3D scans and download 3D models. Available on iPhone, iPad, Android and Web.

Whether you're using C#, Swift, TypeScript, Go, C++ or other languages, quicktype generates models and helper code for quickly and safely reading JSON in your apps. Customize online with advanced options, or download a command-line tool.

All-in-one Website OSINT Scanner

Delegated top-level domains listed and grouped alphabetically.

Your database should be easy to use, not a source of frustration. Xata easily integrates into your developer workflow, providing the best data experience for GitHub, Vercel and Netlify based deployments.

The easiest way to add authentication and user management to your application. Purpose-built for React, Next.js, Remix, and “The Modern Web”.

Add a human touch to your videos with realistic AI avatars. Click to see our selection of 160+ AI avatars across different ethnicities, ages and styles.

HeyGen is an innovative video platform that harnesses the power of generative AI to streamline your video creation process. Unleash your creativity with HeyGen - the future of video production.

Curated weekly analysis of the latest AI projects, products, and news

Learn how to harness the power of AI to boost your productivity at work. Our tutorials show you how to leverage AI for content creation, data analysis, automation and more. Upskill with the latest AI tools and techniques.

Somos la plataforma digital y equipo de contadores que te ayudan a estar siempre al día con el SRI.

Software fiscal para declaración de impuestos del SAT. Conoce Taxo, la mejor plataforma para contadores que automatiza tus procesos en México. ¡Entra ahora!

Easily inject environment variables into your Angular applications. Latest version: 17.3.0-alpha.0, last published: 7 days ago. Start using @ngx-env/builder in your project by running `npm i @ngx-env/builder`. There are no other projects in the npm registry using @ngx-env/builder.

El objetivo de este canal es establecer un punto de referencia centralizado para el contenido de habla hispana en todo lo relacionado con AWS.

[LIVE] Tracking all tech startup layoffs — and lists of employees laid off — since COVID-19 was declared a pandemic. This page is constantly being updated.

AI native in Spanish

Fluid responsive design

🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production - coqui-ai/TTS

GitLab Enterprise Edition

La biblioteca de datos abiertos y código abierto más grande del mundo. Refleja Sci-Hub, Library Genesis, Z-Library, y más.

Playing Pokemon Red with Reinforcement Learning. Contribute to PWhiddy/PokemonRedExperiments development by creating an account on GitHub.

Experience cutting-edge computer vision with our premier image captioning and video summarization algorithms. Tailored for content creators, media professionals, SEO experts, and e-commerce enterprises. Featuring multilingual support and seamless API integration. Elevate your digital presence today.

Internet-scale Neural Networks. Contribute to opentensor/bittensor development by creating an account on GitHub.

why consumer software gets worse, not better, over time

Contribute to omar-florez/congress_ai development by creating an account on GitHub.

Is AI Smart Enough to Summarize Your 500-Page eBook?

https://github.com/BerriAI/litellm

Realtime Diffusion, using Automatic1111 Stable Diffusion API - Fictiverse/Redream

En este tutorial te mostraré paso a paso cómo crear un bot en Discord con todo el conocimiento que tu le entregues mediante la plataforma…

Evaluation tool for LLM QA chains. Contribute to rlancemartin/auto-evaluator development by creating an account on GitHub.

In the world of programming, there are many concepts that every developer should understand in order to build efficient and reliable systems. One such vital concept is idempotency, which refers to the property of an operation or function that produces the same result when applied multiple times as it does

For a long time, humanity has pursued artificial intelligence (AI) equivalent to or surpassing the human level, with AI agents considered a promising vehicle for this pursuit. AI agents are...

Align on your next build faster with Whimsical docs, flowcharts, wireframes, & mind maps. Generate AI diagrams and access templates from product leaders.

AI-first pull request reviewer with context-aware feedback, line-by-line code suggestions, and real-time chat.

Postgres made for developers. Easy to Use, Scalable, Cost efficient solution for your next project.

An open source, cross-platform, customizable ML solution for live and streaming media.

Marqo offers everything needed to run AI search, from training and embedding to fine tuning and retrieval. Get started in minutes, not months.

WarpFusion. Contribute to Sxela/WarpFusion development by creating an account on GitHub.

Lightning-fast, open source search engine for everyone

In this article, we will explore an example of visualizing semantic similarities in language input using OpenAI’s language models and…

Find out if your project name is taken

Learn Deep Learning with fastai and PyTorch, 2022

Drizzle ORM is a lightweight and performant TypeScript ORM with developer experience in mind.

This blog post outlines a key engineering principle I’ve come to believe strongly in for building general AI systems with deep learning. This principle guides my present-day research tastes and day-to-day design choices in building large-scale, general-purpose ML systems. Discoveries around Neural Scaling Laws, unsupervised pretraining on Internet-scale datasets, and other work on Foundation Models have pointed to a simple yet exciting narrative for making progress in Machine Learning: Large amounts of diverse data are more important to generalization than clever model biases. If you believe (1), then how much your model generalizes is directly proportional to how fast you can push diverse data into a sufficiently high-capacity model. To that end, Deep Neural nets trained with supervised learning are excellent data sponges - they can memorize vast amounts of data and can do this quickly by training with batch sizes in the tens of thousands. Modern architectures like ResNets and Transformers seem to have no trouble absorbing increasingly large datasets when trained via supervised learning. When a model has minimized training loss (a.k.a empirical risk), it can be said to have “memorized” the training set. Classically one would think that minimizing training loss to zero is shortly followed by overfitting, but overparameterized deep networks seem to generalize well even in this regime. Here is an illustration of the “double descent” phenomena from Patterns, Predictions, and Actions, which illustrates that in some problems, overparameterized models can continue to reduce test error (risk) even as training loss is fully minimized. A recent ICLR workshop paper investigates this phenomenon on synthetic datasets, showing that if you train long enough in this zero-training-loss regime, the model can suddenly have an epiphany and generalize much later on (the authors call this “Grokking”). Furthermore, the paper also presents evidence that increasing training data actually decreases the amount of optimization required to generalize. It’s as my colleague Chelsea Finn once told me: “Memorization is the first step towards generalization!” State-of-then-art neural networks trained this way can do really impressive things. Here is a DALL-E model that, when prompted with “A banana performing stand-up comedy”, draws the following picture: Here is another DALL-E output, prompted with “an illstration of a baby panda with headphones staring at its reflection in a mirror”. Note that there are no such images of “pandas looking into mirrors” or “banana comedians” in the training data (I think), so these results suggest that the DALL-E model has learned to interpret distinct concepts from text, render the corresponding visual parts in an image and have them interact with each other somewhat coherently. The ability to “just ask” language-conditioned deep learning models for what you want has led to “prompt engineering” as a viable space for improving our ML models. Here is a Tweet discussing how priming a VQGAN + CLIP model with the words “Unreal Engine” leads to drastically higher-quality images. What if we could extend this principle - just asking generalization - to other challenging problems that have eluded analytical algorithmic improvements? Reinforcement Learning: Not a Great Data Sponge In contrast to supervised learning, reinforcement learning algorithms are much less computationally efficient when it comes to absorbing vast quantities of diverse data needed for generalization. To see why this is the case, let’s consider a thought experiment where we train a general-purpose robot to do millions of tasks in unstructured environments. The standard Markov Decision Process is set up as follows: a policy is represented as a state-conditioned distribution over actions, \(p(a \vert s)\), and the environment as consisting of a reward function \(r(s_t, a_t)\) and transition dynamics \(p(s_{t+1} \vert s_t, a_t)\). Initial states and task objectives are encoded in the initial state \(s_0\), which is sampled from a distribution \(p(s_0)\). The goal is to maximize the sum of rewards across the episode, averaged across different starting states sampled from \(p(s_0)\): [\DeclareMathOperator{\argmax}{arg\,max} \DeclareMathOperator{\argmin}{arg\,min} \text{Solve}~\theta^*\ = \argmax_\theta~R(\theta)] [\text{where}~R(\theta)=E_{p(s_0)}[\sum_{t=1}^{T}{r(s_t, a_t)}]~\text{and}~a_t \sim p_\theta(\cdot s_t)~\text{and}~s_{t+1} \sim p(\cdot s_t, a_t)~\text{and}~s_0 \sim p(s_0)] Let’s assume the existence of some optimal policy which we call \(p^\star(a \vert s)\), that achieves the maximum reward \(\max_\theta R(\theta)\). “Supremum” would be more accurate, but I use the \(\max\) operator for notational simplicity. We want to bring our model, \(p_\theta(a \vert s)\), as close as possible to \(p^\star(a \vert s)\). If we had access to the optimal policy \(p^\star(a \vert s)\) as an oracle, we could simply query the oracle action and use it like a supervised learning label. We could then train a feedforward policy that maps the states to the oracle actions, and benefit from all the nice properties that supervised learning methods enjoy: stable training, large batches, diverse offline datasets, no need to interact with the environment. while not converged: batch_states = replay_buffer.sample(batch_size) oracle_actions = [oracle_policy.sample_action(s) for s in batch_states] model.fit(batch_states, oracle_actions) However, in reinforcement learning we often don’t have an expert policy to query, so we must improve the policy from its own collected experience. To do this, estimating the gradient that takes the model policy closer to the optimal policy requires evaluating the average episodic return of the current policy in the environment, and then estimating a gradient of that return with respect to parameters. If you treat the environment returns as a black-box with respect to some parameter \(\theta\) you can use the log-derivative trick to estimate its gradients: [\nabla_\theta E_{p(\theta)} [R(\theta)] = \int_\Theta d\theta \nabla_\theta p(\theta) R(\theta) \ = \int_\Theta d\theta p(\theta) \nabla_\theta \log p(\theta) R(\theta) = E_{p(\theta)} [\nabla_\theta \log p(\theta) R(\theta)]] This gradient estimator contains two expectations that we need to numerically approximate. First is computing \(R(\theta)\) itself, which is an expectation over starting states \(p(s_0)\). In my previous blog post I mentioned that accurate evaluation of a Binomial variable (e.g. the success rate of a robot on a single task) could require thousands of trials in order to achieve statistical certainty within a couple percent. For our hypothetical generalist robot, \(p(s_0)\) could encompass millions of unique tasks and scenarios, which makes accurate evaluation prohibitively expensive. The second expectation is encountered in the estimation of the policy gradient, over \(p(\theta)\). Some algorithms like CMA-ES draw samples directly from the policy parameter distribution \(p(\theta)\), while other RL algorithms like PPO sample from the policy distribution \(p_\theta(a\vert s)\) and use the backpropagation rule to compute the gradient of the return with respect to the parameters: \(\frac{\partial R}{\partial \theta} = \frac{\partial R}{\partial \mu_a} \cdot \frac{\partial \mu_a}{\partial \theta}\). The latter is typically preferred because the search space on action parameters is thought to be smaller than the search space on policy parameters (and therefore requires fewer environment interactions to estimate a gradient for). If supervised behavior cloning on a single oracle label \(a \sim p^\star(a\vert s)\) gives you some gradient vector \(g^\star\), estimating the same gradient vector \(\bar{g} \approx g^\star\) with reinforcement learning requires something on the order of \(O(H(s_0) \cdot H(a))\) times as many episode rollouts to get a comparably low-variance estimate. This is a hand-wavy estimate that assumes that there is a multiplicative factor of the entropy of the initial state distribution \(O(H(s_0))\) for estimating \(R(\theta)\) and a multiplicative factor of the entropy of the action distribution \(O(H(a))\) for estimating \(\nabla_\theta R(\theta)\) itself. Consequently, online reinforcement learning on sparse rewards and diverse, possibly multi-task environments require enormous numbers of rollouts to estimate returns and their gradients accurately. You have to pay this cost on every minibatch update! When the environment requires handling a wide variety of scenarios and demands generalization to unseen situations, it further increases the number of minibatch elements needed. The OpenAI DOTA team found that having millions of examples in their minibatch was required to bring down gradient noise to an acceptable level. This intuitively makes sense: if your objective \(R(\theta)\) has a minimum minibatch size needed to generalize well across many \(s_0\) without excessive catastrophic forgetting, then switching from supervised learning to online reinforcement learning will probably require a larger batch size by some multiplicative factor. What about Offline RL? What about offline RL methods like Deep Q-Learning on large datasets of \((S,A,R,S)\) transitions? These methods work by bootstrapping, where the target values that we regress value functions to are computed using a copy of the same network’s best action-value estimate on the next state. The appeal of these offline reinforcement learning methods is that you can get optimal policies from diverse, off-policy data without having to interact with the environment. Modified versions of Q-learning like CQL work even better on offline datasets, and have shown promise on smaller-scale simulated control environments. Unfortunately, bootstrapping does not mix well with generalization. It is folk knowledge that the deadly triad of function approximation, bootstrapping, and off-policy data make training unstable. I think this problem will only get worse as we scale up models and expect to train them on increasingly general tasks. This work shows that repeated bootstrapping iteratively decreases the capacity of the neural network. If you believe the claim that overparameterization of deep neural networks is key to generalization, then it would appear that for the same neural net architecture, offline RL is not quite as “data absorbent” as supervised learning. In practice, even algorithms like CQL are still challenging to scale and debug on larger, real-world datasets; colleagues of mine tried several variations of AWAC and CQL on large-scale robotics problems and found them to be trickier to get them to work than naive methods like Behavior Cloning. Instead of going through all this trouble, what if we lean into what deep nets excel at - sponging up data quickly with supervised learning and generalizing to massive datasets? Can we accomplish what RL sets out to do using the tools of generalization, rather than direct optimization? Learn the Distribution instead of the Optimum What if we make generalization the first-class citizen in algorithmic design, and tailor everything else in service of it? What if we could simply learn all the policies with supervised learning, and “just ask nicely” for the best one? Consider the recent work on Decision Transformer (DT), whereby instead of modeling a single policy and iteratively improving it with reinforcement learning, the authors simply use supervised learning coupled with a sequential model to predict trajectories of many different policies. The model is conditioned with the Return-to-Go so that it may predict actions consistent with a policy that would achieve those returns. The DT simply models all policies - good and bad - with supervised learning, and then use the magic of deep learning generalization to infer from the expert-conditioned policy. This phenomenon has been observed and developed in several prior and concurrent works, such as Reward-Conditioned Policies, Upside Down Reinforcement Learning and Reinforcement Learning as One Big Sequence Modeling Problem. The AlphaStar team also found that conditioning a model on human player skill level (e.g. future units they ended up build order, MMR, ELO scores) and using it to imitate all player data was superior to only imitating expert-level build orders. This technique is also commonly used in the Autonomous Vehicle space to model both good drivers and bad drivers jointly, even though the autonomous policy is only ever deployed to imitate good driving behavior. Hindsight Language Relabeling At a high level, DTs condition the supervised learning objective on some high level description \(g\) that partitions what the policy will do in the future based on that value of \(g\). The return-to-go is an especially salient quantity for a reinforcement learning task, but you can also express the future outcomes via a goal state or StarCraft build order or even a natural language description of what was accomplished. In Language Conditioned Imitation Learning over Unstructured Data, the authors pair arbitrary trajectories with post-hoc natural language descriptions, and then train a model to clone those behaviors conditioned on language. At test time, they simply “ask” the policy to do a novel task in a zero-shot manner. The nice thing about these techniques is that they are indispensable for reaching sparse goals on RL tasks like Ant-Maze. This lends support to the claim that generalization and inference across goal-conditioning can do far better than brute force search for a single sparse goal in a long-horizon task. Language is a particularly nice choice for conditioning because it can be used to partition a trajectory not just on skill level, but also by task, by how much the policy explores, how “animal-like” it is, and any other observations a human might make about the trajectory. Clauses can be composed ad-hoc without developing a formal grammar for all outcomes that the robot might accomplish. Language is an ideal “fuzzy” representation for the diversity of real-world outcomes and behaviors, which will become increasingly important as we want to partition increasingly diverse datasets. Generalizing From Imperfect Demonstrations A recent work I am quite inspired is D-REX, which tackles the problem of inferring the environment’s reward function from the demonstrations of a suboptimal policy. Classically, one requires making an assumption that the demonstrator is the optimal policy, from which you can use off-policy algorithms (e.g. Q-learning) to estimate the value function. Offline value estimation with deep neural nets can suffer from poor generalization to state-action pairs not in the demonstrator trajectory, and thus requires careful algorithmic tuning to make sure that the value function converges. An algorithm with poor convergence properties makes the propsects of minimizing training loss - and therefore generalization - tenuous. D-REX proposes a really clever trick to get around not having any reward labels at all, even when the demonstrator is suboptimal: Given a suboptimal policy \(\pi_\theta\), generate trajectory rollouts \(\tau_1, \tau_2, ... \tau_N\) by having the policy interact with the environment. On each rollout, add variable amounts of noise \(\epsilon\) to its actions. Assume that adding noise to a suboptimal policy makes it even more suboptimal, i.e. \(R(\tau) \geq R(\tau + \epsilon)\). Train a ranking model \(f_\theta(\tau_i, \tau_j)\) to predict which of two trajectories \(\tau_i, \tau_j\) has a higher return. The ranking model magically extrapolates to trajectories that are better than what \(\pi_\theta\) can generate, even though the ranking model has never been trained on trajectories better than \(\pi_\theta\) itself. I like this approach because ranking models are stable to train (they are just classifiers), and this method is able to achieve better-than-demonstrator behavior not through the explicit construction of the Bellman inequality or implicit planning through a learned model, but rather via extrapolation on a family of perturbations. Do You Even Need RL to Improve from Experience? In the above sections I’ve described how you can “generalize and infer” to get around exploration and even inverse reinforcement learning from sparse rewards. But what about “improving from a policy’s own experience, tabular rasa”? This is the main reason why people put up with the pain of implementing RL algorithms. Can we replace this with supervised learning algorithms and a bit of generalization as well? The goal of RL is to go from the current set of parameters \(\theta^{n}\) and some collected policy experience \(\tau\) to a new set of parameters \(\theta^{n+1}\) that achieves a higher episode return. Instead of using a “proper” RL algorithm to update the agent, could we just learn this mapping \(f: (\theta^{n}, \tau) \to \theta^{n+1}\) via supervised deep learning? This idea is sometimes referred to as “meta-reinforcement learning”, because it involves learning a better reinforcement learning function than off-the-shelf RL algorithms. My colleagues and I applied this idea to a project where we trained a neural network to predict “improved policy behavior” from a video of a lesser policy’s experience. I could imagine this idea being combined with ranking and trajectory augmentation ideas from D-REX to further generalize the “policy improvement behavior”. Even if we never train on optimal policy trajectories, perhaps sufficient data augmentation can also lead to a general improvement operator that extrapolates to the optimal policy regime of parameters. People often conflate this policy improvement behavior with “reinforcement learning algorithms” like DQN and PPO, but behavior is distinct from implementation. The “policy improvement operator” \(f: (\theta^{n}, \tau) \to \theta^{n+1}\) can be learned via your choice of reinforcement learning or supervised learning, but is deployed in a RL-like manner for interacting with the environment. The “Just-Ask-Generalization” Recipe Here is a table summarizing the previously mentioned RL problems, and comparing how each of them can be tackled with a “generalize-and-infer” approach instead of direct optimization. Goal “Direct Optimization” Approach “Generalize + Inference” Approach Reinforcement Learning with Sparse Rewards Find \(p^\star(a_t\vert s_t)\) s.t. \(R_t=1\), brute force exploration DT: Learn \(p(a_t\vert s_t,R_t)\) from many policies, infer \(p(a_t\vert s_t, R_t=1)\). H.E.R - Infer tasks for which gathered trajectories are optimal, then learn \(p(\text{trajectory}\vert \text{task})\). Then infer optimal trajectory for desired task. Learn a Reward Function from Suboptimal Trajectories Offline Inverse RL D-REX: Trajectory augmentation + Extrapolate to better trajectories. Improve the policy from experience Q-Learning, Policy Gradient Watch-Try-Learn: Learn \(p(\theta^{n+1} \vert \theta^n , \tau, \text{task})\) Fine-tune a simulated policy in a real-world environment Sample-efficient RL fine-tuning Domain Randomization: train on a distribution of simulators, and the policy “infers which world” it is in at test time. The high-level recipe is simple. If you want to find the solution \(y_i\) for a problem \(x_i\), consider setting up a dataset of paired problems and solutions \((x_1, y_1), ..., (x_N, y_N)\) and then training a deep network \(y = f_\theta(x)\) that “simply maps your problems to solutions”. Then substitute your desired \(x_i\) and have the deep network infer the solution \(y_i\) via generalization. “Problem” is meant in the most abstract of terms and can refer to a RL environment, a dataset, or even a single example. “Solutions” could be represented as the optimal parameters of a policy or a neural network, or a single prediction. Techniques like goal relabeling help generate post-hoc problems from solutions, but building such a dataset can also be achieved via data augmentation techniques. At its core, we are transforming a difficult optimization problem into an inference problem, and training a supervised learning model on a distribution of problems for which it’s comparatively cheap to obtain solutions. To summarize the recommendations in a three-step recipe: Choose a method capable of minimizing training loss on massive datasets, i.e. supervised learning with maximum likelihood. This will facilitate scaling to complex, diverse datasets and getting the most generalization mileage out of your compute budget. If you want to learn \(p(y\vert x, \text{task}=g^\star)\) for some prediction task \(g^\star\), try learning \(p(y\vert x, \text{task})\) for many related but different tasks \(g \sim p(g), g \neq g^\star\) Then at test time just condition on \(g^\star\). Formulate conditioning variables that help partition the data distribution while still admitting generalization on held-out samples from \(p(g)\). Natural language encoding is a good choice. The insight that we can cast optimization problems into inference problems is not new. For example, the SGD optimizer can be cast as approximate Bayesian inference and so can optimal control via AICO. These works present a theoretical justification as to why inference can be a suitable replacement for optimization, since the problems and algorithms can be translated back and forth. I’m suggesting something slightly different here. Instead of casting a sequential decision making problem into an equivalent sequential inference problem, we construct the “meta-problem”: a distribution of similar problems for which it’s easy to obtain the solutions. We then solve the meta-problem with supervised learning by mapping problems directly to solutions. Don’t overthink it, just train the deep net in the simplest way possible and ask it for generalization! Perhaps in the near future we will be able to prompt-engineer such language-conditioned models with the hint “Generalize to unseen …”. Just ask for … Consciousness? How far can we stretch the principle of “generalize-and-infer” as an alternative to direct optimization? Here is a “recipe for consciousness” which would probably be better pondered over some strong drinks: Train a language-conditioned multi-policy model \(p_\theta(a\vert s, g)\) (implemented via a Decision Transformer or equivalent) to imitate a variety of policies \(\pi_1, ..., \pi_N\) conditioned on natural language descriptions \(g\) of those agents. At test time, some default policy \(p(a\vert s, g=\text{Behave as myself})\) interacts with another agent \(\pi_\text{test}\) for a number of steps, after which we instruct the model to “behave as if you were \(\pi_\text{test}\).” The model would require a sort of “meta-cognition of others” capability, since it would have to infer what policy \(\pi_\text{test}\) would do in a particular situation. We make a copy of the multi-policy model \(p_\phi \sim p_\theta\), and embed multiple test-time iterations of step (1) within a single episode, with dozens of agents. Two of these agents are initially conditioned as \(p_\theta(a\vert s, g=\text{Behave as myself})\) and \(p_\phi(a\vert s, g=\text{Behave as myself})\). This generates episodes where some agents imitate other agents, and all agents observe this behavior. Then we ask \(p_\phi\) to emit actions with the conditioning context “behave as if you were \(\pi_\theta\) pretending to be you”. This would require \(\pi_\phi\) to model \(\pi_\theta\)’s imitation capabilities, as well as what information \(\pi_\theta\) knows about \(\pi_\phi\), on the fly. Researchers like Jürgen Schmidhuber have previously discussed how dynamics models (aka World Models) of embodied agents are already “conscious”, because successful modeling the dynamics of the environment around oneself necessitates a representation of the self as an embodied participant in the environment. While I think that “self-representation” is a necessity in planning and dynamics prediction problems, I think the framework is too vacuous to be of use in reproducing a convincing imitation of consciousness. After all, any planning algorithm that represents “the self” explicitly within each imagined trajectory rollout would be conscious under this definition. An A* maze-planner would satisfy this definition of consciousness. What I’m proposing is implementing a “more convincing” form of consciousness, not based on a “necessary representation of the self for planning”, but rather an understanding of the self that can be transmitted through language and behavior unrelated to any particular objective. For instance, the model needs to not only understand not only how a given policy regards itself, but how a variety of other policies might interpret the behavior of a that policy, much like funhouse mirrors that distort one’s reflection. The hypothesis is that through demonstrating this understanding of “distorted self-reflection”, the policy will learn to recognize itself and model the internal motivations and beliefs of other agents in agent-agent interactions. There are some important implementation details that I haven’t fleshed out yet, but at high level, I do think that supervised learning and natural language conditioning with enormous agent-interaction datasets are sufficiently powerful tools to learn interesting behaviors. Imbuing agents with some kind of meta-cogition ability of the self and other agents is an important step towards a convincing imitation of consciousness. Acknowledgements Thanks to Daniel Freeman, David Ha, Karol Hausman, Irwan Bello, Igor Mordatch, and Vincent Vanhoucke for feedback and discussion on earlier drafts of this work. Citation If you want to cite this blog post, you can use: @article{jang2021justask, title = "Just Ask for Generalization", author = "Jang, Eric", journal = "evjang.com", year = "2021", month = "Oct", url = "https://evjang.com/2021/10/23/generalization.html" } References Generalization and scaling: Scaling Laws for Neural Language Models Self-supervised Pretraining of Visual Features in the Wild On the Opportunities and Risks of Foundation Models Understanding deep learning requires rethinking generalization A Large Batch Optimizer Reality Check: Traditional, Generic Optimizers Suffice Across Batch Sizes Patterns, Predictions, Actions: Generalization Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets DALL·E: Creating Images from Text RL challenges: Robots Must Be Ephemeralized An Empirical Model of Large-Batch Training Implicit Under-Parameterization Inhibits Data-Efficient Deep Reinforcement Learning Deep Reinforcement Learning and the Deadly Triad Conservative Q-Learning AW-Opt: Learning Robotic Skills with Imitation andReinforcement at Scale Hindsight Imitation Decision Transformer: Reinforcement Learning via Sequence Modeling Reward-Conditioned Policies Upside Down Reinforcement Learning Reinforcement Learning as One Big Sequence Modeling Problem Grandmaster level in Starcraft II via multi-agent reinforcement learning Hindsight Experience Replay Learning Latent Plans from Play Replacing RL with Supervised Learning Better-than-Demonstrator Imitation Learning via Automatically-Ranked Demonstrations Watch, Try, Learn: Meta-Learning from Demonstrations and Rewards Distribution Augmentation for Generative Modeling Stochastic Gradient Descent as Approximate Bayesian Inference Robot Trajectory Optimization using Approximate Inference Q/A Igor Mordatch supplied interesting questions and comments in reviewing this blog post. I have paraphrased his questions here and added responses in this section. 1. You discussed Supervised Learning and Reinforcement Learning. What do you think about Unsupervised Learning and “The Cake Analogy”? I consider unsupervised learning to be simply supervised learning for a different task, with comparable gradient variance, since targets are not usually noisly estimated beyond augmentation. Maximum likelihood estimation and contrastive algorithms like InfoNCE seem to be both useful for facilitating generalization in large models. 2. For the first difficulty of RL (evaluating success), aren’t there parallels to current generative models too? Success evaluation is hard for language models, as evidenced by dissatisfaction with BLEU scores and difficulty of evaluating likelihoods with non-likelihood based generative image models. There are parallels to likelihood-free generative models which require extensive compute for either training or sampling or likelihood evaluation. In practice, however, I think the burdens of evaluation are not directly comparable, since the computational expense of marginalization over observations for such models is dwarfed by the marginalization of success rate estimation in RL. In RL, you have to roll out the environment over O(coin flips) x O(initial state distribution) x O(action distribution) in order to get a low-variance policy gradient for “improved success across all states and tasks”. O(coin flips) is O(1000) samples for local improvement of a couple percent with statistical certainty, wheras I think that typically the marginalization costs of implicit likelihood tends to be cheaper with tricks like Langevin sampling O(minibatch=32). Also, the backprop passes used in Langevin dynamics are usually cheaper than running full environment simulations with a forward pass of the neural net on every step. 3. One of the findings of current language model work is that proxy objectives for what you really want are good enough. Simple next-token prediction induces generalization. But alignment to what you really want is still a hard problem in large model field and we don’t have good answers there yet (and ironically many attempts so far relied on incorporation of RL algorithms). Alignment objectives may lack a per-example surrogate loss. But under the “generalize-then-infer” school of thought, I would simply recommend learning \(p(y\vert x, \text{alignment objective})\) with max likelihood over numerous hindsight alignment objectives, and then simply condition on the desired alignment object at test time. One could obtain a distribution of alignment descriptions by simply running the model live, and then hindsight labeling with the corresponding alignment realized by the model. Then we simply invoke this meme by Connor Leahy: Just asking the AI to be nice sounds flippant, but after seeing DALL-E and other large-scale multi-modal models that seem to generalize better as they get bigger, I think we should take these simple, borderline-naive ideas more seriously. 4. For the second difficulty of RL (gradient estimation), we know that for settings where you can backprop through environment dynamics to get exact policy gradient, doing so often leads to worse results. This reminds me of an old FB comment by Yann Lecun that a better way to estimate Hessian-vector products with ReLU activations is to use a stochastic estimator rather than computing the analytical hessian, since the 2nd-order curvature of ReLU is 0 and what you actually want is the Hessian-vector product of the smoothed version of the function. If you need to relax the dynamics or use an unbiased stochastic estimator to train through a differentiable simulator, then I think you’re back to where you’re starting with expensive evaluation, since presumably you need many rollouts to smooth out the simulator function and reduce variance. However, maybe the number of samples you need to estimate a smoothed policy gradient is a reasonable tradeoff here and this is a nice way to obtain gradients. 5. Why hasn’t something as simple as what you propose (generalize-then-infer) been done already? Some researchers out there are probably pursuing this already. My guess is that the research community tends to reward narratives that increase intellectual complexity and argue that “we need better algorithms”. People pay lip service to “simple ideas” but few are willing to truly pursue simplicity to its limit and simply scale up existing ideas. Another reason would be that researchers often don’t take generalization for granted, so it’s often quicker to think about adding explicit inductive biases rather than thinking about generalization as a first-class citizen and then tailoring all other design decisions in support of it. 6. How does your consciousness proposal relate to ideas from Schmidhuber’s “consciousness in world models” ideas, Friston’s Free Energy Principle, and Hawkin’s “memory of thoughts”? I consider Schmidhuber and Friston’s unified theories as more or less stating “optimal control requires good future prediction and future prediction with me in it requires self-representation”. If we draw an analogy to next-word prediction in large language models, maybe optimizing next state prediction perfectly is sufficient for subsuming all consciousness-type behaviors like theory-of-mind and the funhouse self-reflections I mentioned above. However, this would require an environment where predicting such dynamics accurately has an outsized impact on observation likelihoods. One critique I have about Schmidhuber and Friston’s frameworks is that they are too general, and can be universally applied to sea slugs and humans. If a certain environmental complexity is needed for future prediction to give rise to something humans would accept as conscious, then the main challenge is declaring what the minimum complexity would be. Hawkin’s “consciousness as memory of perception” seems to be more related to the subjective qualia aspect of consciousness rather than theory of mind. Note that most people do not consider a program that concatenates numpy arrays to be capable of “experiencing qualia” in the way humans do. Perhaps what is missing is the meta-cognition aspect - the policy needs to exhibit behaviors suggesting that it contemplates the fact that it experiences things. Again, this requires a carefully designed environment that demands such meta-cognition behavior. I think this could emerge from training for the theory-of-mind imitation problems I described above, since the agent would need to access a consistent representation about how it perceives things and transform it through a variety of “other agent’s lenses”. The flexibility of being able to project one’s own representation of sensory observations through one’s representation of other agents’ sensory capabilities is what would convince me that the agent understands that it can do sufficient meta-cognition about qualia. 7. Your formulation of consciousness only concerns itself with theory-of-mind behavior. What about attention behavior? See the second paragraph of the response to #6. Update 20211025: Updated with a paraphrased question from Alexander Terenin 8. In Rich Sutton’s Bitter Lesson Essay, he argues that search and learning are both important. Do you really think that search can be completely replaced by a learned approach? I agree that having a bit of light search in your program can be immensely helpful to learning and overall performance. It’s a bit of a chicken/egg though. Does AlphaGo work because MCTS uses a learned value function to make search tractable? Or does the policy distillation only work because of search? I’m suggesting that when search becomes too hard (most RL tasks), it’s time to use more learning. You’re still doing search when performing supervised learning - you just get a lot more gradient signal per flop of computation.

AI Observability & Evaluation. Contribute to Arize-ai/phoenix development by creating an account on GitHub.

An open-source & self-hostable Heroku / Netlify / Vercel alternative.

AllMusic provides comprehensive music info including reviews and biographies. Get recommendations for new music to listen to, stream or own.

Quadratic is an infinite canvas spreadsheet with AI, Python, and SQL - all running in your web browser.

En el fascinante mundo del procesamiento de lenguaje natural, las herramientas para transformar documentos y dividir textos se han vuelto…

Temporary disposable numbers to receive SMS online for accounts verification and QA tests

Who's built ChatGPT for teams? ◆ Saved and categorized chats◆ Realtime prompt co-editing◆ System prompt templates◆ Integrations (Drive, Dropbox, etc)◆ "Chat with documents"◆ ACL & RBAC◆ BYO models and full privacy control🤔— Guillermo Rauch (@rauchg) July 23, 2023

A smooth scroll library fresh out of the Studio Freight Darkroom

Python bindings for the Transformer models implemented in C/C++ using GGML library. - marella/ctransformers

AI-Exchange Protocol (AIXP): A Communication Standard for Artificial Intelligence Agents - davila7/AIXP

Design and develop React projects visually on one development platform. With Codux, your team can collaborate faster to create better UI.

In this post, Phillip talks through the challenges & pitfalls of LLMs we faced when building our Query Assistant - and that you too may face.

Doctor Dignity is an LLM that can pass the US Medical Licensing Exam. It works offline, it's cross-platform, & your health data stays private. - llSourcell/Doctor-Dignity

Free open source Tailwind CSS starter templates and components to get you started quickly to creating websites in Tailwind CSS!

An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena. - lm-sys/FastChat

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

'Herika - The ChatGPT Companion' is a revolutionary mod that aims to integrate Skyrim with Artificial Intelligence technology. It specifically adds a follower, Herika, whose responses and interactions

Get answers to complex questions with Phind's AI answer engine.

A reference architecture for the LLM app stack. It shows the most common systems, tools, and design patterns used by AI startups and tech companies.

Aprende a utilizar la Inteligencia Artificial en tu día a día, en Castellano.

Free, local and privacy-aware chatbots

Head to Integrations for documentation on built-in integrations with 3rd-party vector stores.

I run the "npx dalai alpaca install 7B" command, and cmd seems to just stop, without downloading the model. Anyone have any ideas? Thanks!

428 TV reviews. Start with our picks for the best TVs below. These are the ones we recommend based on our testing and in-depth TV reviews.

Find Inspiration in Design Elements.

Built with Notion, the all-in-one connected workspace with publishing capabilities.

Rix is a developer-focused search engine powered by AI. You can use the Web-search mode to find the latest information or switch to GPT-knowledge for instant answers from pre-trained knowledge.

A hub of integrations for LlamaIndex including data loaders, tools, vector databases, LLMs and more.

Use powerful AI apps on FlowGPT, the largest AI platform, for free! Get instant answers from characters, resume editor, essay generator, coding wizard, and more!

Without a clear presentation, leaders will ignore you. When you create presentations with STORYD, leaders pay attention and you get recognized for your hard work.

Visla is an AI powered video creation platform that enables teams and individuals to easily record, edit, and share high-quality videos.

¿Pensando en hacer compras? ahorra dinero y tiempo al comparar los precios de las mejores tiendas del Perú en un solo lugar.

Discover amazing ML apps made by the community

Create audio like writing a doc. Edit words not waveforms, switch speakers and tweak pronunciations with phonetics.

E-mails, subdomains and names Harvester - OSINT . Contribute to laramies/theHarvester development by creating an account on GitHub.

AI to enhance, upscale and generate pictures. Increase resolution and quality. Fix blurry, pixelated, low res images. Make every photo sharp and clear. Start for free.

GPT-4 is OpenAI’s most advanced system. GPT-4 can solve difficult problems with greater accuracy, thanks to its broader general knowledge and problem solving abilities. It can generate, edit, and iterate with users on creative and technical writing tasks, such as composing songs, writing screenplays, or learning a user’s writing style.

Discover amazing ML apps made by the community

Sprout Social's research reveals network-specific recommendations for the best times to post on LinkedIn globally and across industries.

Además, 58% de los profesores no cuentan con un título en educación . Desde el 2020, el Indecopi ha sancionado a más de 500 centros educativos privados por distintas irregularidades.

Spline is a free 3D design software with real-time collaboration to create web interactive experiences in the browser. Easy 3d modeling, animation, textures, and more.

PlayHT is #1 AI Voice Generator with 600+ AI voices that creates ultra realistic Text to Speech voiceovers. Convert text to audio and download as MP3 & WAV files.

Developer blogging with custom domains, hosted/headless CMS options. Our new headless CMS streamlines content management for devtool companies.

A Python package with a built-in web application

Futurepedia is a free site to help you find the best AI tools and software to make your work and life more efficient and productive. Updated daily, join millions of followers of our website, newsletter and YouTube.

Fast, reliable, and secure dependency management.

You can run yarn commands on another path with the --cwd flag yarn --cwd /path/to/command add new-package yarn --cwd is equivalent...

A Bearer Token provides information to an API request about an account from a trusted authority. Azure Active Directory can serve as a trusted authority to authenticate an account and provide a Bearer Token in JSON Web Token (JWT) format. This video shows how to accomplish this.

You my want change your WordPress URL, this allows you to change the WordPress domain/subdomain, enable or disable “www”, etc. this process can be done by multiple ways but in this guide we will tell you how to change it using phpMyAdmin, we will avoid using third party plugins by doing so.

Get the best continuous integration and delivery (CI/CD) for any platform, in our cloud or on your own infrastructure, for free.

In this post, we will upload a file into Google Drive using the Google Drive API in a Node.js application, create a folder, and list files.

Friendly tutorials for developers. Focus on React, CSS, Animation, and more!

Get more out of your generative AI workflows without breaking the bank

Head to Integrations for documentation on built-in integrations with document loader providers.

Learn more about the different methods used to search for moonrise-replaceefe688efbbef4d9db29ba180cce56ee2moonrise-replace files and folders.

Plugin para WordPress que permite generar y mejorar descripciones automaticas para productos de WooCommerce, con conexión a ChatGPT (OpenAI).

Use ChatGPT with enhanced features like chat history search, folders, integrations, prompt library, etc.

Artículo original: How to Remove Images and Containers in Docker [https://www.freecodecamp.org/news/how-to-remove-images-in-docker/] Traducido y adaptado por: Sil Zubikarai [/espanol/news/author/sil/] Docker rmi docker rmi remueve imágenes por su  ID. Para remover la imagen, primero necesitas una lista de todas las imágenes para obtener las IDs de imagen, nombre de la imagen, y

A graphic design app that helps you create professional quality social media posts, invitations, digital postcards, graphics, and more. Start with your idea and create something unique for you.

A new medium for presenting ideas, powered by AI. Create beautiful, engaging content with none of the formatting and design work.

Perplexity AI unlocks the power of knowledge with information discovery and sharing.

AI powered video editing platform to repurpose long form podcasts and videos into short viral clips for TikTok, Instagram Reels and YouTube Shorts.

Create royalty-free, background music with the power of AI. Recommended for videos, podcasts & games. Try now!

Enhance and edit photos with Fotor’s free online photo editor. Crop, add text & effects, retouch images, and more using powerful photo editing tools.

What are autonomous agents? Why are they such a big opportunity? How do they work? What does this look like in the future? How can I build or use one?

Pitch Deck Hunt is a curation of the best startup pitch decks. Discover the best pitch deck examples from unicorns to new startups!

Open Source backend in 1 file with realtime database, authentication, file storage and admin dashboard

Assemble, configure, and deploy autonomous AI Agents in your browser.

Forget lengthy development cycles and expensive machine learning teams. With Synthflow you can build sophisticated, tailored AI agents without technical skills or coding - just bring your data and ideas.

Transform your ideas to videos effortlessly with the best AI Video Generator. User-friendly Text to Video editor, realistic voiceovers, dynamic AI clips, and more.

Transform your projects with our AI image generator. Generate high-quality, AI generated images with unparalleled speed and style to elevate your creative vision

Learn Prompting is the largest and most comprehensive course in prompt engineering available on the internet, with over 60 content modules, translated into 9 languages, and a thriving community.

Bundlephobia helps you find the performance impact of npm packages. Find the size of any javascript package and its effect on your frontend bundle.

Una plataforma abierta que te permite gestionar tu carrera online y mantener SIEMPRE el control sobre tu CV

Create your own ML-powered opera song! by David Li with Google Arts & Culture

Adrenaline: Your Expert AI Programming Assistant. Get instant help with coding questions, debug issues, and learn programming. Ideal for developers and students alike.

LangChain is a framework for developing applications powered by language models. It enables applications that:

3116 Free SVG icons for popular brands

Se trata de la reciente sentencia dictada por el Juzgado Civil Transitorio de San Juan de Miraflores, con competencia en asuntos de familia, a cargo del magistrado Frank Paul Flores García, en el Expediente 00052-2022-18-3002-JP-FC-01, sobre proceso de alimentos.

AI-accelerated music production, in your browser. This is the next-generation DAW.

Create beautiful images from your tweets in a snap

Build & Share Delightful Machine Learning Apps.

ChatGPT for Research. Consensus is an AI-powered search engine that finds and summarizes scientific research papers. Just ask a question!

Build production-grade applications with a Postgres database, Authentication, instant APIs, Realtime, Functions, Storage and Vector embeddings. Start for free.

The ChatGPT Retrieval Plugin lets you easily find personal or work documents by asking questions in natural language. - openai/chatgpt-retrieval-plugin

Community driven roadmaps, articles and guides for developers to grow in their career.

A collective list of free APIs. Contribute to public-apis/public-apis development by creating an account on GitHub.

AppFlowy is an AI-powered secure workspace where you achieve more without losing control of your data

A list of SaaS, PaaS and IaaS offerings that have free tiers of interest to devops and infradev - ripienaar/free-for-dev

Learn about developer use cases using Node, React, Python, and more. Explore developer docs, new features and tech content.

Next.js 13.2 introduces major improvements to the App Router in preparation for stability, including SEO support, Route Handlers, MDX for Server Components, Type-Safe Links, and more.

Projects launch grant programs to reward teams and individuals to build in their ecosystem. Blockworks is compiling a real-time directory list of all grants, RFPs, and Bug Bounties. Apply here for available crypto grant programs.

Free AI image generator. Free AI art generator. Free AI video generator. 100+ models and styles to choose from. Train your personalized model. Most popular AI apps: sketch to image, image to video, inpainting, outpainting, model fine-tuning, real-time drawing, text to image, image to image, image to text and more!

BLACKBOX AI is the Best AI Model for Code. Millions of developers use Blackbox Code Chat to answer coding questions and assist them while writing code faster. Whether you are fixing a bug, building a new feature or refactoring your code, ask BLACKBOX to help. BLACKBOX has real-time knowledge of the world, making it able to answer questions about recent events, technological breakthroughs, product releases, API documentations & more BLACKBOX integrates directly with VSCode to automatically suggests the next lines of code based on your repo context.

Write code quickly and efficiently with AI code suggestions

Explore developer resources, tutorials, API docs, and dynamic examples to get the most out of OpenAI's platform.

AI Video Generation

AI Voice Generator in 20 languages. 120+ realistic text to speech voices to create the perfect AI voiceover. Go instantly from text to voice with ease.

Concise, consistent, and legible badges

Download thousands of free and premium Illustrations. Instantly customize them to fit your brand colors. Use them in your designs, apps, marketing collaterals and more.

Free illustrations you can use for your design projects. Click to download hundreds of beautiful vector illustrations for free today!

The design project with open-source illustrations for any idea you can imagine and create. Create beautiful websites, products and applications with your color, for free.

Mix-&-match illustrations of people with a design library for InVIsion Studio and Sketch.

Free and premium vector SVG illustrations for you to use on your next project, no attribution required! Vector illustrations, icons, 3D illustrations, animation, mockups, Figma plugins and more.

Large set of opensource illustrations. You can use completely free and without attribution.

Create Fly Volumes from 1 to 500 gigabytes with a single command; they show up as directories in your VMs, use them however you like.

Railway is an infrastructure platform where you can provision infrastructure, develop with that infrastructure locally, and then deploy to the cloud.

Algorithm Visualizer is an interactive online platform that visualizes algorithms from code.

An interactive Git visualization tool to educate and challenge!

Our philosophy to scaling is simple, avoid premature optimization

Convert any workflow into a step-by-step guide in seconds, automatically

Unlock AI insights for your files instantly. Ask, learn, and extract data 10X faster with Humata.

Framer is where teams design and publish stunning sites.

Free Vectors and Icons in SVG format. ✅ Download free mono or multi color vectors for commercial use. Search in 500.000+ Free SVG Vectors and Icons.

You provide a link to your knowledge base. Fini takes care of the rest. Get a tireless AI agent, ready to answer customer questions instantly, 24/7. Integrates with Intercom, Search, Slack and Discord. All without code.

Restore your old photos and keep the memories alive.

Guest Post

Krisp cancels background noise, records, transcribes, and summarizes meetings and calls.

Conversational and continuously learning, You.com enhances web search, writing, coding, digital art creation, and solving complex problems.

Puedes descargar la última versión del navegador Brave con bloqueo de anuncios y rastreadores aquí. Prepárate para replantearte por completo cómo funciona internet.

A blog featuring tutorials about JavaScript, HTML5, AJAX, PHP, CSS, WordPress, and everything else development.

Tabnine is an AI code assistant that makes you a better developer | Trusted by millions of developers | Supports all programming languages and IDEs

DeepL Write is a tool that helps you perfect your writing. Write clearly, precisely, with ease, and without errors. Try for free now!

Railway is an infrastructure platform where you can provision infrastructure, develop with that infrastructure locally, and then deploy to the cloud.

Always here to listen and talk. Always on your side. Join the millions growing with their AI friends now!

Drag and drop with your components, right within your existing site or app. Build and optimize digital experiences for any tech stack, visually..

There are simple podcast & video editors and there are powerful ones. Only Descript is both & it features magical AI, so you can skip the hard part of editing. Get started for free.

An online Sankey diagram builder for everyone

Render is a unified cloud to build and run all your apps and websites with free TLS certificates, global CDN, private networks and auto deploys from Git.

Automatic code review by AI

We’ve trained and are open-sourcing a neural net called Whisper that approaches human level robustness and accuracy on English speech recognition.

New research shows that just 0.01% of bitcoin holders controls 27% of the currency in circulation.

Shie Mannor and Aviv Tamar

Dramatically simplify API integrations with Tray connectors, the Connector Builder, and our Authentication and Projects features. Visit Tray.io to learn more.

Glide makes it easy to build and deploy powerful custom apps powered by AI with no code. Create your first app by trying Glide for free.

Discover the future of software development with Bubble, a no-code platform used to build SaaS tools, marketplaces, CRMs, & more — hosted securely in the cloud.

Softr turns your Airtable or Google Sheets data into beautiful and powerful client portals or internal tools. Pick a template or start from scratch.

Insights is a project by Get on Board that reports live, up-to-date trends of the tech talent market in Latin America, based on our unique data on thousands of jobs, applications, and professionals.

A modern, simple, and blazingly fast way to collaborate – bring knowledge, docs, and projects together in one place.

Dive into teamwork with templates. Ready-to-use docs for product specs, company handbooks, meeting notes and more.

Stories from the future of the web.

Cryptocurrency (Bitcoin, Ethereum etc) blockchain live transaction visualizer. Every tx is a person, and they fill up buses which represent blocks.

A nice collection of often useful examples done in React.js

Antler presents the first comprehensive map of the Generative AI landscape available to everyone.

Ionic empowers web developers to build leading cross-platform mobile apps and Progressive Web Apps (PWAs)

From beautiful event pages to effortless invites and ticketing, Luma is all you need to host a memorable event.

Twilio Segment is a customer data platform (CDP) that helps you collect, clean, and activate your customer data.

Runway is an applied AI research company shaping the next era of art, entertainment and human creativity.

Track your web3, crypto and DeFi portfolio in one dashboard. Easily monitor and analyze the activity of multiple wallets across 45+ EVM and non-EVM chains.

Sinónimos Gratis: Diccionario de sinónimos, antónimos, y palabras relacionadas. Sinónimos Gratis contiene más de 31.000 entradas únicas desde fuentes principales incluyendo Diccionario Manual de Sinónimos y Antónimos de la Lengua Española. Busque palabras o frases para encontrar sinónimos, antónimos, y palabras relacionadas, todos destacados por color y agrupados por significado. Filtre mediante categoría gramatical para mostrar sólo los resultados que usted desea ver. Encuentre rápidamente los sinónimos para decenas de miles de palabras con Sinónimos Gratis.

Directorio de comunidades tecnológicas de habla hispana. Conoce comunidades por temas y regiones.

Free Online Image Editor create your own animated gifs resize crop avatars and images. Photo tool for your favorite pictures. Edit an image here fast and easy online.

Change the size of PNG pictures and photos. Online PNG img size reducer.

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Kaggle is the world’s largest data science community with powerful tools and resources to help you achieve your data science goals.

Generador de códigos QR gratuito para crear códigos QR ✅ uso gratuito y comercial ✅ sin fecha de caducidad ► Crea tus códigos QR gratuitos ahora

Explore developer resources, tutorials, API docs, and dynamic examples to get the most out of OpenAI's platform.

Collection of templates for you to use in your own forms.

We’ve trained a model called ChatGPT which interacts in a conversational way. The dialogue format makes it possible for ChatGPT to answer followup questions, admit its mistakes, challenge incorrect premises, and reject inappropriate requests.

Cuando empiezas en desarrollo web es muy útil tener algunas herramientas o páginas a la mano que te permitan ser más eficiente al momento de crear

Instructions on how to create a GPT3-powered business emails. Designed for people with literacy challenges or written communication skills.

LanguageTool es un servicio de revisión de textos en inglés, español/castellano, catalán/valenciano, francés y 30 idiomas más. Revisa al instante los errores de gramática y estilo de tu texto.

Customize Torus Wallet UI, branding, and translations to fit your brand

Easily build and share beautiful presentations with Pitch. Start for free on pitch.com

the place to find and be found. summer 2024.

Torus is the most secure passwordless authentication and private key management platform with the security guarantees of non-custodial Public Key Infrastructure (PKI).

Web3Auth is a simple, non-custodial auth infrastructure that enables Web3 wallets and applications to provide seamless user logins to both mainstream and native Web3 users.

Where web3.0 happens

How we approached a web3 UX problem and helped design a unique airdrop

Sprout reveals the best times to post on social media for peak engagement. Armed with this data is a game changer to your social strategy.

Photosphere Viewer for React.JS, wrapper of photo-sphere-viewer.js. Latest version: 5.0.2-psv5.7.1, last published: a month ago. Start using react-photo-sphere-viewer in your project by running `npm i react-photo-sphere-viewer`. There are 3 other projects in the npm registry using react-photo-sphere-viewer.

A library of resources to help you share content, grow your audience, and build an independent subscription business in the creator economy!

Learn about Alchemy's Road to Web3 series, a 10-week, self-paced web3 developer program to help new blockchain devs go from beginner to advanced.

A powerful, lightweight theme for Hugo built with Tailwind CSS.

Myriad.Town stands at the crossroads of real life and digital interactions.

RealityChain is an award-winning social metaverse that allows users to generate, deploy and customize decentralized metaverses easily.

Want to know what Shopify theme a store is using? Just use our Shopify theme detector and find out what theme is being used by a Shopify store

Intro I’ve been using my setup for working with customers and projects in Obsidian for a while, and thought I wanted to share how Obsidian can be used in a work environment. It uses templates and scripts a lot to automate processes. My work is within sales, and so I have many customer meetings as well as customer projects I’m working on. The reason for posting here is for other people finding inspiration on how to setup their own systems, with every use case of course being unique. This is wha...

We're testing digital collectibles on Instagram so you can show off your NFTs.