Skip to content
Active Inference Ontology
Correct & Incorrect Examples
Active Inference Ontology ~ Correct & Incorrect Examples
List
Term
Correct Examples
Incorrect Examples
Core
65
Accuracy
A lower-resolution camera or fMRI has lower and thus lesser capacity to map fine scale features of stimuli.
@Accuracy
The true distribution was 10 +/- 1 and the person guessed 10 +/- 5, thus they had higher . I was sure the green car was really blue; this inaccuracy was caused by the dim light.
@Accuracy
Action
Ants are continually involved in in real life () and/or in a ()
@Action
@Realism
@Model
@Instrumentalism
I love this band’s first song, can’t wait till the really begins!
@Action
Action Planning
The robot assessed its current and target location, then engaged in to decide under time pressure how to navigate.
@Action Planning
The used algorithms for on
@Agent
@Action Planning
@Variational Free Energy
@Inference
@Sensory input
Active Inference
It is fun and rewarding to learn and apply as a framework for , , and — it is a compatible with the .
@Active Inference
@Perception
@Cognition
@Action
@Process Theory
@Free Energy Principle
is a style of reinforcement learning that uses statistics only.
@Active Inference
@Artificial
@intelligence
@frequentist
Active States
The of the computer program are the statistical outputs that it presents, while the are the incoming statistical dependencies.
@Active States
@Sensory input
@Sense State
Canada and Kenya have many people involved in dance, really you could say these two are
@Active States
Affordance
“In this T-maze model, there are 3 for movement at the junction (Left, Right, Down).”
@Affordance
“He didn’t have the to do that, but he did it anyway”.
@Affordance
Agency
An uses their to sculpt their which makes it more predictable thereby minimizing the agent’s and maintaining its in the face of random fluctuations.
@Agent
@Agency
@environment
@Variational Free Energy
@Markov Blanket
I got a new picture taken at the passport .
@Agency
Agent
“The ant nestmate is the in the Active InferAnts model”
@Agent
“Calcium carbonate is the anti-caking in this cookie”
@Agent
Ambiguity
The noisy readings from the thermometer resulted in high given the to the .
@Ambiguity
@Sensory input
@Agent
I am curious about what is inside the box, there is about what it might be.
@Ambiguity
Attention
The instantly updated to the new because it was paying maximal to the .
@Model
@Sensory Data
@Attention
@Stimulus
The words were impactful even though the child was not paying at all.
@Attention
Bayesian Inference
Many of the key ideas of existed before Rev. Bayes, and in some cases reflect recent contributions from computational research.
@Bayesian Inference
The frequentist t-test is a classical example of .
@Bayesian Inference
Behavior
I expect you to be on your best !
@Behavior
The of this hammer is being determined entirely by my s without being intermediated by s.
@Behavior
@Internal State
@Blanket State
Belief
The or in can be a on .
@Prior
@Empirical prior
@Bayesian Inference
@Belief
@Hidden State
A in requires or conscious awareness.
@Belief
@Bayesian Inference
@Subjective feeling states
Belief updating
The incoming resulted in
@Sensory Data
@Belief updating
I promise that if you vote for me, even if I change my mind due to new coming in, I will never undergo — so you know exactly what you get with me!
@Information
@Belief updating
Blanket State
I don’t care whether it is an or , as long as it is a !
@Action
@Sense State
@Blanket State
My child created a fortress from the bedding; it belongs to the
@Blanket State
Cognition
BacillisAxelis345 engaged in to decide whether to eat or to escape.
@Cognition
Because the is exhibiting intelligent , this means
@Agent
@Behavior
Complexity
can refer to the number of predictor or independent variables or features that a model needs to take into account in order to make accurate prediction.
@Model
@Complexity
is the opposite of concavity.
@Complexity
Data
These three readings from the thermometer constitute !
@Data
An anecdote is not a piece of .
@Data
Ensemble
The ant colony from a al, or perspective, is an of nestmates.
@Behavior
@Collective behavior
@Ensemble
After the band played their final song, the crowd called out “!”, prompting them to play another song.
@Ensemble
Epistemic value
When the environment contains I would like to undertake the action of for the sake of it to see if I can gather more relevant s to explain what is going on around me. Such relevant s have for me as an .
@Uncertainty
@Foraging
@Observation
@Observation
@Epistemic value
@Agent
In the case of temperature homeostasis, taking s that move the body temperature in an adaptive direction are of purely .
@Action
@Epistemic value
Evidence
Every photon is like a piece of on the retina.
@Evidence
Just because you have litigated the case with doesn’t mean that you provided any
@Data
@Evidence
Expectation
At timestep 1, the made a prediction about through time, this was an about the future.
@Agent
@Expected Free Energy
@Expectation
Although the scientist made a about the future using a , it was not an since they were not waiting for it to be realized.
@Prediction
@Model
@Expectation
Expected Free Energy
The deep affective used calculation as a basis of .
@Inference
@Agent
@Expected Free Energy
@Policy selection
The amount of calories that the has at is the .
@Living system
@Non-Equilibrium Steady State
@Expected Free Energy
External State
In my of temperature, the true temperature in the room is being modeled as an (), so we will never know it.
@Generative Model
@External State
@Hidden State
That country is outside of our borders, it is an .
@External State
Free Energy
can refer to various different formulations and decompositions, that share some important key features.
@Free Energy
You can get the electricity for no cost at all, it is .
@Free Energy
Free Energy Principle
As a principle, the cannot be falsified.
@Free Energy Principle
There is not enough empirical evidence to support the .
@Free Energy Principle
Friston Blanket
A is a type of where s are associated with incoming and states are associated with outgoing of the .
@Friston Blanket
@Markov Blanket
@Sense State
@Sensory Data
@Action
@Behavior
@Agent
Karl walked into the room and cast a chilly over the audience.
@Friston Blanket
Generalized Free Energy
wrote “Crucially, this means the reduces to the for outcomes that had been observed in the past.....Outcomes in the formulation are represented explicitly as s. This means that the over is incorporated explicitly in the generative model.”
@Generalized Free Energy
@Variational Free Energy
@Generalized Free Energy
@Belief
@Prior
@Sensory outcome
The speaker was talking about on in a very broad and vague fashion, this is identical to .
@Variational Free Energy
@Generalized coordinates
@Generalized Free Energy
Generative Model
es are commonly used for computational modeling of s.
@Partially Observed Markov Decision Process
@Generative Model
The subject of the photography session made art as well, they are a — but if they wouldn’t have made the art, they would have only been a , not a .
@Generative Model
@Model
@Generative Model
Generative Process
The generates .
@Generative Process
@Sensory Data
The uses its to determine future of the .
@Agent
@Generative Process
@Hidden State
@environment
Hierarchical Model
A is a , for example the .
@Nested
@Model
@Hierarchical Model
@Hierarchically Mechanistic Mind
The stated that water boils at a higher temperature at higher altitudes.
@Hierarchical Model
Inference
The researcher made of where the was doing on ( , ) as well as ( ).
@Model
@Active Vision
@Agent
@Inference
@Action
@Action Planning
@Action Prediction
@Perception
@perceptual inference
The ball rolled downhill using on , sometimes veering more to the left and other times more to the right.
@Inference
@Policy selection
Information
There is more maximum in 1 terabyte than in 1 gigabyte.
@Information
I listened to that podcast and the file checksum was OK but the was modified relative to the reference version.
@Information
Internal State
When the describes a forest outside, and the s describe the bark of the tree, the s describe the core of the tree.
@Generative Process
@Blanket State
@Internal State
The of the internet is equivalent to the total number of computers connected to it at any given time.
@State space
Learning
The software agent engaged in on internal s, this is technically .
@Belief updating
@parameter
@Learning
Every day we change, but from a perspective it is only if the is adaptive.
@Bayesian Inference
@Learning
@Belief updating
Living system
A body is a .
@Living system
A is a mathematical model used to predict weather patterns.
@Living system
Markov Blanket
s are features of Maps (e.g. s), not of Territories (e.g. ant colonies or brains).
@Markov Blanket
@Bayesian
@Graphical
@Model
In textile industry, a is a blanket made using a specific knitting technique."
@Markov Blanket
Markov Decision Process
Whether it is fully observable or not, one common type of model used in control theory is a
@Markov Decision Process
We will let Captain Markov make the call here, since it is a
@Markov Decision Process
Multi-scale system
The of counties within states within countries, was a
@Generative Model
@Multi-scale system
In culinary arts, a is a method for measuring ingredients in both metric and imperial units.
@Multi-scale system
Niche
Every ant lives in their ecological .
@Niche
That band’s kind of music honestly is just too for me.
@Niche
Non-Equilibrium Steady State
Blood sugar levels are dynamic and fluctuating, however they revisit characteristic states repeatedly, this can be described by a .
@Non-Equilibrium Steady State
The ball just lay still on the floor, we can describe this as exhibiting a .
@Non-Equilibrium Steady State
Observation
One can make all the difference.
@Observation
s are not required in order to do empirical parametrization of a statistical model.
@Observation
Particle
and s together constitute the .
@Internal State
@Blanket State
@Particle
The house was made of board.
@Particle
Perception
Visual gives us many demonstrations of the characteristics of our — for example saccades, the blind spot, and blink supression.
@Perception
@Generative Model
models and as literally the same thing for each .
@Active Inference
@Action
@Perception
@Agent
Policy
The sequence of s that an plans to take is a .
@Action
@Agent
@Policy
What insurance do you recommend for my house?
@Policy
Policy selection
When the decided to go one way instead of the other, it was due to an internal process of , specifically .
@Agent
@Policy selection
@Action and Planning as Divergence Minimization
We went shopping for insurance and the package was already determined for me, this is a case of .
@Stochastic
@Policy selection
Posterior
The distribution reflects our degrees of about s after we see .
@Posterior
@Belief
@Latent cause
@Sensory Data
The distribution can always be trivially obtained by solving Bayes’ theorem.
@Posterior
Pragmatic Value
The had s for there to be more beans in the jar, so adding more beans was of for them.
@Agent
@Preference
@Pragmatic Value
If a reduces the divergence between your s and s, it is defined as having negative .
@Policy
@Preference
@Observation
@Pragmatic Value
Prediction
After successfully the structure of the the can make a about the future state of this and the associated it will generate.
@Learning
@Generative Process
@Agent
@Prediction
@Generative Process
@Sensory Data
A magic 8-ball can make a that will reliably be true.
@Prediction
Preference
The of the bacteria underwent fitting ( / ) on , guided by a for medium but not high/low sugar concentration.
@Generative Model
@parameter
@Belief updating
@Learning
@Action
@Preference
All results in s that realize their in terms of .
@Action
@Agent
@Preference
@Sensory outcome
Prior
uses on .
@Active Vision
@Prior
@Sensory input
We arrested someone with no s.
@Prior
Recognition Model
After constructing a , an can invert this model to obtain the which allows for the prediction of the (causes) that generated some new .
@Generative Model
@Agent
@Recognition Model
@Hidden State
@Sensory Data
That billboard has a picture of one of the most in the world.
@Recognition Model
Representation
The sterotypical neural pattern induced by a is considered a , at least by those who subscribe to .
@Stimulus
@Representation
@Representationalism
No Taxation without !
@Representation
Salience
is related to how relevant a given appears to be.
@Salience
@Stimulus
I can smell dinner cooking, and already my mouth is .
@Salience
Sense State
In this model, the s coming in from the thermometers are considered as s.
@Observation
@Sense State
Most models don’t show it, but actually all s include a “sixth “.
@Active Inference
@Generative Model
@Sense State
State
is a type of that partition from
@Blanket State
@State
@Internal State
@External State
California is the with the best honey on the West Coast.
@State
State space
The of a can be a or .
@State space
@Generative Model
@Continuous state space
@Discrete state space
In politics, the refers to the total area governed by a particular state.
@State space
Stationarity
A common assumption of many time series algorithms is that the data exhibits .
@Stationarity
I am exhibiting when I stop walking.
@Stationarity
Surprise
Surprisal cannot be minimized directly because it is involves calculating the term in Bayes’ Rule which generally involves an intractable integral over all possible states an organism can be in.
@Evidence
When they walk into the room, yell ““!
@Surprise
System
George Mobus argues that s have some fundamental properties such as structure and function.
@System
A is the scientific term for a large group of wolves.
@System
Temporal Depth
In deep temporal models, or occurs because of the number of possibilities one must account for increases as more future states are modeled (see: )
@Temporal Depth
@Counterfactual
The longer the queue, the greater the .
@Temporal Depth
Uncertainty
A measure of unpredictability or expected (cf, ). The about a is often quantified with its (inverse ).
@Surprise
@Entropy
@Uncertainty
@Random variable
@Variance
@Precision
I felt a lot of after that job interview.
@Uncertainty
Variational Free Energy
is a tractable way to compute an upper bound on of a given .
@Variational Free Energy
@Surprisal
@Generative Model
@Data
Our electric bill fluctuates so much each month though on average it is zero; this is known as .
@Variational Free Energy
Outcome
The produces outcomes when we sample from it.
@Generative Process
I was worried about the of my decision.
@Outcome
Habit
Supplement
301
Abstract Action
Abstract action prediction
Abstract Bayesian Inference
Abstract epistemic value
Abstract External State
Abstract Generative Model
Abstract Hidden State
Abstract Internal State
Abstract Sensory State
Abstract System
Abstract Accuracy
abstractCounterpart
Action and Planning as Divergence Minimization
Action at a distance
Action Integral
Action Prediction
The over the next few timesteps with respect to , is the .
@Generative Model
@Active States
@Action Prediction
The inferred what s it had, this process is known as or .
@Agent
@Affordance
@Action Planning
@Action Prediction
Active Vision
@Active Vision
Agency based model
Agency free model
Algorithm
Alignment
analogy
Appraisal theories of emotion
Attenuation of response
Augmented reality
Autopoiesis
In the right niche, cells can be considered to exhibit at the level.
@Autopoiesis
@System
The pile of sand quickly dissipated in the wind, however I still think it is my favorite example of long-range .
@Autopoiesis
Bayes-optimal control
Bayesian
Bayesian Brain
Bayesian surprise
Bethe approximation
Bottom-up attentional control
chaos
Circular causality
Cognitive Science
Cognitive System
Cognitivism
Collective behavior
Conceptual metaphor
Conditional Probability
Confidence
Congruence
Connectionism
Control (states)
Control theory
Counterfactual
Cue
Culture
Cybernetics
Decision-making
Density
Deontic Action
Development
Dissipation
Distribution
divergence
Kullback-Leibler Divergence
Domain
Domain-generality
Domain-specificity
Dynamic Causal Modelling
Dynamic expectation maximization
Dynamicism
Ecology
EcoEvoDevo
Embedded Embodied Encultured Enactive Inference
Embodied Cybernetic Complexity
Embodied Belief
Emotion
Empirical prior
Enactivism
Entropy
Ergodicity
Estimator
Event-related potential
Evolution
Expectation maximization
Expected Utility Theory
Experience of body ownership
Explaining Away
Explanation
Extended Cognition
Factor graph
Falsification
Far-from-equilibrium
Fokker-Planck Equation
Foraging
Forney
Friction
Friston's Law
functional magnetic resonance imaging (FMRi)
Gaussian distribution
Generalized coordinates
Generalized Synchrony
Generative density
Generative modelling
Gestalt
Goal-driven selection
Gradient Descent
Graphical
Group Renormalization Theory
Guidance signal
Habit learning/formation
Hamilton's Principle of Least Action
Helmholtz machine
Hidden State
Hierarchically Mechanistic Mind
Homeostasis
Homeostatic system
Hyperprior
Hypothesis
Information bottleneck
Information Geometry
Instrumentalism
Interface
Interoception
Interoceptive sensitivity
Interpretation
Inverse problem
Latent cause
Lateral geniculate nucleus
Likelihood
Marginal approximation
Markovian Monism
Marr's Levels of Description
Material science
Mean
Mean field approximation
Memory
Message Passing
Mismatch negativity
Mode
Model
Model accuracy
Model Inversion
Morphogenesis
Multisensory integration
Narrative
I wrote a about my time in graduate school.
@Narrative
Network
Neuronal Ensemble
Niche construction
Noisy signal
Non-linear dynamical systems
Novelty
Candy and ice cream are kinds of foods.
@Novelty
Optimal control
overfitting
Partition
Policy posterior
Policy prior
Precision
Prediction error
Prediction error minimization
Predictive Coding
Predictive Processing
Principle
Probability distribution
Process Theory
Friston “The distinction is between a [theory] and ; i.e., the difference between a normative principle that things may or may not conform to, and a or hypothesis about how that principle is realized”
@State
@Process Theory
@Process Theory
Random variable
Realism
Receptive field
Recognition density
Regime of Attention
Representationalism
Reservoir Computing
Reward
Risk
Sample space
Selection bias
Selection history
Self-organization
Selfhood
Semi-Markovian
Sense of agency
Sensorimotor
Sensory attenuation
Sensory Data
Sensory input
Sensory outcome
Sentience
Shared Generative Model
Signal
Simulation
solenoidal
Sophisticated Inference
spike-timing dependent plasticity
Statistical manifold
Stigmergy
Stochastic
Subjective feeling states
Surprisal
Swarm
Symbol
Synergetics
Teams
Theory
Thermodynamic system
Thermostatistics
Thing
Thinking Through Other Minds
Top-down attentional control
Umwelt
Unidirectionality
Update
Variance
Variational
Variational Niche Construction
Von Economo neurons
Weak mixing
Working memory
World States
Active learning
Attracting set
Bayesian belief updating
Bayesian mechanics
Coarse graining
Continuous state space
Discrete state space
Epistemic foraging
Hermeneutics
Least action
link
Partially Observed Markov Decision Process
Renormalization
Variational message passing
Variational principle
Lagrangian
Path integral
Sensory observation
changing mind (cognition)
changing world (action)
High road
The to the starts by talking about random in general without a specific focus on biological organisms with brains.
@High road
@Free Energy Principle
@Non-linear dynamical systems
You take the and I’ll take the low road.
@High road
Low road
The to the starts by looking at how biological organisms perceive their and take actions within it to develop a notion about how they can successfully predict the next state they will be in ( as testing).
@Low road
@Free Energy Principle
@environment
@Perception
@Hypothesis
Be careful, the can be dangerous.
@Low road
normative
active processes
changing the mind
changing the world
model evidence
predictive machine
Bayesian Model Selection
Blanket index
categorical
constraint
fitness
flow
Gauge theory
Helmholtz Free Energy
maximum caliber
path
phenotype
population
Quantum
Quantum-like
Qubit
Sensory State
Sensory states
sufficient statistic
time
Active Blockference
Analytical Philosophy
Category Theory
Continental Philosophy
DeSci
Exteroception
Filter
Helmholtz Decomposition
Proprioception
Quantum mechanics
Statistical Parametric Mapping
T-Maze
Effective
Effectivity
Nested
Path of Least Action
Precariousness
affect
Deflationary
Inflationary
Co-category
Functor
Lens
Monad
anticipation
matching
tracking
category
colimit
morphism
Allostasis
Artificial
frequentist
intelligence
Natural
sparsity
tensor network
Bayes Theorem
Credibility
degrees of freedom
interaction
tensor
[7] is a single-dimensional in 1 dimension.
@tensor
unit of adaptive behavior
Gain
Gain is related to and
@Precision
@Variance
Entailed
74
Want to print your doc?
This is not the way.
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.