[New] Concise and Practical AI/ML

Pages
- Preface
- Artificial Intelligence
  Concepts
  High-level Intelligence
- Maths for ML
  Calculus
  Algebra
- Machine Learning
  History of ML
  ML Models
  ML Model is Better
  How a Model Learns
  Boosted vs Combinatory
  Neuralnet
  Neuron
  Types of Neurons
  Layers
  Neuralnet Alphabet
  Heuristic Hyperparams
  Feedforward
  Input Separation
  Backprop
  Activation Functions
  Loss Functions
  Gradient Descent
  Optimizers
  Design Techniques
  Normalization
  Regularization
  Drop-out Technique
  Concatenation
  Overfitting & Underfitting
  Explosion & Vanishing
  Engineering Techniques
- Methods of ML
  Supervised Learning
  Regression
  Classification
  Reinforcement Learning
  Concepts
  Bellman Equation
  Q-table
  Q-network
  Learning Tactics
  Policy Network
  Unsupervised Learning
  Some Applications
  Other Methods
- Practical Cases
- Ref & Glossary

[New] Concise and Practical AI/ML

Ref & Glossary

Explore

Ref & Glossary

Books

Deep Learning —

https://www.deeplearningbook.org⁠

⁠

G4G ML —

https://www.geeksforgeeks.org/machine-learning⁠

⁠

Libraries

TensorFlow —

https://www.tensorflow.org/tutorials⁠

⁠

Backend for training

Keras —

https://www.tensorflow.org/api_docs/python/tf/keras⁠

⁠

High-level APIs utilising backend

HuggingFace —

https://huggingface.co⁠

⁠

High-level Toolset for training

Trained model cloud storage

Tools

Colab —

https://colab.research.google.com⁠

Models

Gemma —

https://huggingface.co/google⁠

⁠

Kaggle —

https://www.kaggle.com⁠

⁠

Trained model cloud storage

Servers with GPU

Others

Communities

Plairo —

http://plairo.novaeh.com⁠

Blogs

Danzwork —

https://danzwork.wordpress.com⁠

Glossary

https://developers.google.com/machine-learning/glossary⁠

⁠

Neuron — A math-based neuron simulate a little bit of bio-neuron

Layer — A chain of neurons usually function with the same context (input for output of previous layer)

Q-learning — A method of Reinforcement Learning using Bellman q-update (based on Bellman equation), Q-learning can use either q-table or q-network

Q-table — The original knowledge store used in q-learning, huge memory needed

Q-network — The modern knowledge store (params) in q-learning, very much memory efficient, as it learns by param combination instead of slots for all cases

Q-update — The formula for updating Q-value based on Bellman equation, Q += Rate x T, where T is temporal difference.

Q-value — Estimated sum of rewards from a state and taking a specific action

Q-function — The function which returns q-value, can be the q-table or q-network

Policy Network — A network in RL which returns suggested action instead of q-value

AI — Artificial Intelligence, with learning or no learning

ML — Machine Learning, the process of learning new things in AI

DL — Deep Learning, the method of ML using artificial neural network (ANN, neuralnet)

Bias — The param associated with each neuron to shift the separation line defined by weights

Weight — The param in a neuron for modifying input to neuron, and together with other weights and biases it makes the combination to comprehend left-most input

Loss Function — A convex function (parabola pointing downward) to minimize

Want to print your doc?
This is not the way.

Try clicking the ··· in the right corner or using a keyboard shortcut (

CtrlP

) instead.