Active Inference Institute Act. Infer. Serve.
ZH

此页面由机器翻译成中文。 查看英文原版。

知识工程师、研究人员以及专注于“主动推断”的正式知识表示领域的逻辑学家。

主动推断本体

维护和发展“Active Inference”公共本体及去中心化科学。

Ontology shortlinkRepository

在本页

主动推断本体 指南

最佳下一步行动

主动推断本体路径

从本页最高信号的公开链接开始，然后继续浏览相关资源和目录视图。

Ontology shortlinkResearchRepositoryRepository过滤资源全球目录教育活跃单元项目

积极推断本体项目维护并扩展了积极推断框架的公共本体。它提供结构化、机器可读的概念及其关系定义，支持去中心化科学、可复现性和知识再利用。

参与

欢迎知识论师、知识工程师以及从事正式知识表示的研究者贡献。描述逻辑、RDF/OWL或概念建模方面的背景有帮助。

Discord inviteCommunityEduActive Unit

图视图

The ontology as a connected concept map

Every term in the Active Inference Ontology, colored by tag, with the connections declared between them. Select a concept to highlight what it links to; the full definitions are in the table below.

Jump to the full ontology terms table

Ontology terms

All 429 Active Inference Ontology terms

The full public ontology (v5): each term with its tag, definition, and a correct example. The graph above shows how the terms connect.

429 terms across 8 tags

Active Inference Ontology (v5) terms with tags, definitions, and examples.
TermTagDefinitionExample
AccuracyBayesian StatisticsBroad sense: how “close to the mark” an Estimator is.A lower-resolution camera or fMRI has lower Accuracy and thus lesser capacity to map fine scale features of stimuli.
AmbiguityBayesian StatisticsBroad sense: Extent to which stimuli have multiple plausible interpretations, requiring priors &/or Action for disambiguationThe noisy readings from the thermometer resulted in high Ambiguity given the Sensory input to the Agent .
Bayesian InferenceBayesian StatisticsAs opposed to frequentist analysis, Bayesian Inference uses a specified Prior or Empirical prior to Update the distributional PosteriorMany of the key ideas of Bayesian Inference existed before Rev. Bayes, and in some cases reflect recent contributions from computational research.
BeliefBayesian StatisticsBroad sense: Felt sense by an Agent of something being true, or confidence it is the case.The Prior or Empirical prior in Bayesian Inference can be a Belief on Hidden State .
Belief updatingBayesian StatisticsBelief updating is changes in a Bayesian Inference Belief through time.The incoming Sensory Data resulted in Belief updating
ComplexityBayesian StatisticsThe extent to which an Agent must revise a Belief to explain incoming Sensory observations.Model 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.
DataBayesian StatisticsData are a set of values of qualitative or quantitative variables about one or more Agent or object .These three readings from the thermometer constitute Data!
ExpectationBayesian StatisticsWithin a Bayesian Inference framework, Expectation is an Estimator about future timestepsAt timestep 1, the Agent made a prediction about Expected Free Energy through time, this was an Expectation about the future.
InferenceBayesian StatisticsProcess of reaching a (local or global) conclusion within a Model, for example with Bayesian Inference.The researcher made Model of Active Vision where the Agent was doing Inference on Action (Action Planning , Action Prediction ) as well as Perception (perceptual inference ).
InformationBayesian StatisticsMeasured in bits, the reduction of Uncertainty on a Belief distribution of some type. Usually Syntactic (Shannon) but also can be Semantic (e.g. Bayesian ).There is more maximum Information in 1 terabyte than in 1 gigabyte.
LearningBayesian StatisticsBroad sense: Process of an Agent engaged in Updates to Cognition (and possibly) Behavior.The software agent engaged in Belief updating on internal parameters, this is technically Learning .
OutcomeBayesian StatisticsIf we consider the environment as a Generative Process that can be sampled when in a particular state, the statistical result (Data, Sensory observation) of the sampling is known as the outcome.The Generative Process produces outcomes when we sample from it.
PosteriorBayesian StatisticsThe Update to the Prior after Observation has occurredThe Posterior distribution reflects our degrees of Belief about Latent causes after we see Sensory Data.
PredictionBayesian StatisticsAn Estimator about a State in a Model at a future time.After successfully Learning the structure of the Generative Process the Agent can make a Prediction about the future state of this Generative Process and the associated Sensory Data it will generate.
PriorBayesian StatisticsThe initial or preceding state of a Belief in Bayesian Inference, before Sensory Data (Observation or Evidence ) occurs.Active Vision uses Prior on Sensory input .
StateBayesian Statisticsis the statistical, computational, or mathematical value for a parameter within the State space of a Model .Blanket State is a type of State that partition Internal State from External State
State spaceBayesian StatisticsSet of variables/parameters that describe a System .The State space of a Generative Model can be a Continuous state space or Discrete state space .
StationarityBayesian StatisticsOf a Random variable , that it is described by parameters that are drawn from a Gaussian distribution and unchanging over the time horizon of analysis.A common assumption of many time series algorithms is that the data exhibits Stationarity.
SurpriseBayesian StatisticsIn Bayesian Inference, Surprise is the negative log evidence, directly corresponding to the inverse of probability (high probability, low surprisal; low probability, high surprisal). The proxy that bounds surprisal is the difference between Prior and Posterior Distribution — how “surprising” Sensory Data are to the Generative Model of the Agent.Surprisal cannot be minimized directly because it is involves calculating the Evidence term in Bayes’ Rule which generally involves an intractable integral over all possible states an organism can be in.
Temporal DepthBayesian StatisticsThe length of a time window or horizon considered (longer time → deeper / more Temporal Depth )In deep temporal models, Temporal Depth or occurs because of the number of Counterfactual possibilities one must account for increases as more future states are modeled (see: https://www.frontiersin.org/articles/10.3389/fpsyg.2018.00579/full)
UncertaintyBayesian StatisticsIn Bayesian Inference , a measure of the Expectation of Surprise (Entropy) of a Random variable (associated with its variance or inverse Precision )A measure of unpredictability or expected Surprise (cf, Entropy ). The Uncertainty about a Random variable is often quantified with its Variance (inverse Precision ).
ActionActionBroad sense: The dynamics, mechanisms, and measurements of BehaviorAnts are continually involved in Action in real life (Realism) and/or in a Model (Instrumentalism)
Action PlanningActionThe selection of an Affordance based upon Inference of Expected Free EnergyThe robot assessed its current and target location, then engaged in Action Planning to decide under time pressure how to navigate.
AgencyActionThe ability of an Agent to engage in Action in their Niche and enact Goal-driven selection or Policy selection based upon PreferenceAn Agent uses their Agency to sculpt their environment which makes it more predictable thereby minimizing the agent’sVariational Free Energy and maintaining its Markov Blanket in the face of random fluctuations.
BehaviorActionThe sequence of Action that an Agent is observed to enact.I expect you to be on your best Behavior! The Active Inference Agent uses its selected Policy to take Action that we call its Behavior.
PolicyActionSequence of Actions, reflected by series of Active States as implemented in Policy selection which is Action Prediction or Action and Planning as Divergence MinimizationThe sequence of Actions that an Agent plans to take is a Policy.
Policy selectionActionThe process of an Agent engaging in Action Planning from set of Affordance, in Active Inference based upon minimization of Expected Free EnergyWhen the Agent decided to go one way instead of the other, it was due to an internal process of Policy selection, specifically Action and Planning as Divergence Minimization.
PreferenceActionparameter in Bayesian Inference Markov Decision Process that ranks or scores the extent to how an Agent values Sensory input .The Generative Model of the bacteria underwent parameter fitting (Belief updating / Learning ) on Action , guided by a Preference for medium but not high/low sugar concentration.
Active InferenceFree EnergyActive Inference is a Process Theory related to Free Energy Principle .It is fun and rewarding to learn and apply Active Inference as a framework for Perception, Cognition , and Action — it is a Process Theory compatible with the Free Energy Principle.
Epistemic valueFree Energyis the value of Information gain or Expectation of reduction in Uncertainty about a State with respect to a Policy, used in Policy selectionWhen the environment contains Uncertainty I would like to undertake the action of Foraging for the sake of it to see if I can gather more relevant Observations to explain what is going on around me. Such relevant Observations have Epistemic value for me as an Agent.
Expected Free EnergyFree EnergyMeasure for performing Inference on Action over a given time horizon (Policy selection , Action and Planning as Divergence Minimization ).The deep affective Inference Agent used Expected Free Energy calculation as a basis of Policy selection .
Free EnergyFree EnergyFree Energy is an Information Theoretic quantity that constitutes an upper bound on Surprisal .Free Energy can refer to various different formulations and decompositions, that share some important key features.
Free Energy PrincipleFree EnergyA generalization of Predictive Coding (PC) according to which organisms minimize an upper bound on the Entropy of Sensory input (or sensory signals) (the Free Energy). Under specific assumptions, Free Energy translates toPrediction error .As a principle, the Free Energy Principle cannot be falsified.
Generalized Free EnergyFree EnergyPast Variational Free Energy plus future Expected Free Energy (each totaled over Policy).Parr and Friston 2019 wrote “Crucially, this means the Generalized Free Energy reduces to the Variational Free Energy for outcomes that had been observed in the past.....Outcomes in the Generalized Free Energy formulation are represented explicitly as Beliefs. This means that the Prior over Sensory outcome is incorporated explicitly in the generative model.”
Pragmatic ValueFree EnergyPragmatic Value is the benefit to an organism of a given Policy or Action , measured in terms of probability of a Policy leading to Expectation of Random variable values that are aligned with the Preference of the AgentThe Agent had Preferences for there to be more beans in the jar, so adding more beans was of Pragmatic Value for them.
Variational Free EnergyFree EnergyMeasure that performs e.g. Inference on Sensory Data given a Generative Model , F. The two sources that composeVariational Free Energy are Model overfitting and Model accuracyVariational Free Energy is a tractable way to compute an upper bound on Surprisal of a Generative Model given Data.
Active StatesMarkov PartitioningIn the Friston Blanket formalism, the Blanket State are the Sense State (incoming Sensory input) and Active States (outgoing influence of Policy selection )The Active States of the computer program are the statistical outputs that it presents, while the Sensory input Sense State are the incoming statistical dependencies.
Blanket StateMarkov PartitioningSet of states in the Markov Blanket Partition that make Internal State and External State have Conditional Probability that are independent.I don’t care whether it is an Action or Sense State, as long as it is a Blanket State!
External StateMarkov PartitioningStates with Conditional density independent from Internal State, conditioned on Blanket State.In my Generative Model of temperature, the true temperature in the room is being modeled as an External State (Hidden State), so we will never know it.
Friston BlanketMarkov PartitioningMarkov Blanket with partitioned Active States and Sense State .A Friston Blanket is a type of Markov Blanket where Sense States are associated with incoming Sensory Data and Action states are associated with outgoing Behavior of the Agent.
Internal StateMarkov PartitioningStates with Conditional density independent from External State , conditioned on Blanket State.When the Generative Process describes a forest outside, and the Blanket States describe the bark of the tree, the Internal States describe the core of the tree.
Markov BlanketMarkov PartitioningMarkov Partitioning Model of System, reflecting Agent as delineated from the Niche via an Interface. The Markov Blanket Blanket State reflect the State(s) upon which Internal State and External State are conditionally independent.Markov Blankets are features of Maps (e.g. Bayesian Graphical Models), not of Territories (e.g. ant colonies or brains).
Markov Decision ProcessMarkov PartitioningBayesian Inference Model where Agent Generative Model can implement Policy selection on Affordances reflected by Active States, while other features of the Generative Process are outside the Control (states) of the Agent .Whether it is fully observable or not, one common type of model used in control theory is a Markov Decision Process
Sense StateMarkov PartitioningIn the Friston Blanket formalism, the Blanket State are the Sense State (incoming Sensory input) and Active States (outgoing influence of Policy selection )In this model, the Observations coming in from the thermometers are considered as Sense States.
AffordanceAgents in the NicheOptions or capacities for Action by an Agent (sometimes called “Affordance 3.0”)“In this T-maze model, there are 3 Affordance for movement at the junction (Left, Right, Down).”
AgentAgents in the NicheEntity as modeled by Active Inference , with Internal State separated from External State by Blanket State“The ant nestmate is the Agent in the Active InferAnts model”
CognitionAgents in the NicheAn Agent modifying the weights of its Internal State for the purpose of Action Planning and/or Belief updating . (This is a @realistCounterpart of Goal-driven selection .)BacillisAxelis345 engaged in Cognition to decide whether to eat or to escape.
EnsembleAgents in the NicheGroup of more than one Agent.The ant colony from a Behavioral, orCollective behavior perspective, is an Ensemble of nestmates.
Generative ModelAgents in the NicheA formalism that describes the mapping between Hidden State, and Expectations of Action Prediction , Sensory outcome .Partially Observed Markov Decision Processes are commonly used for computational modeling of Generative Models.
Generative ProcessAgents in the NicheUnderlying @dynamical process in the Niche giving rise to Agent Observation and @agent Action PredictionThe Generative Process generates Sensory Data.
NicheAgents in the NicheEcology System constituting the Generative Process (as Partitioned from the Agent who instantiates a Generative Model ).Every ant lives in their ecological Niche.
Non-Equilibrium Steady StateAgents in the NicheTechnically, a Non-Equilibrium Steady State requires a solution to the Fokker Planck equation (i.e., density dynamics). A nonequilibrium steady-state solution entails solenoidal (i.e., conservative or divergence free) dynamics that break detailed balance (and underwrite stochastic chaos ). In other words, The dynamics of systems at Non-Equilibrium Steady State are not time reversible (unlike equilibrium steady states, in which the flow is entirely dissipative).Blood sugar levels are dynamic and fluctuating, however they revisit characteristic states repeatedly, this can be described by a Non-Equilibrium Steady State.
ParticleAgents in the NicheAn Agent consisting of Blanket State and Internal State , partitioned off from Niche .Internal State and Blanket States together constitute the Particle.
Recognition ModelAgents in the NicheRecognition Model is the kind of Model that affords Variational Inference, which lets us calculate or approximate a probability distribution. Recognition Model is a synonym for Variational Model.After constructing a Generative Model, an Agent can invert this model to obtain the Recognition Model which allows for the prediction of the Hidden State (causes) that generated some new Sensory Data.
RepresentationAgents in the NicheA structural correspondence between some Random variable inside a System and some Random variable outside the System (isomorphism being the strongest kind of correspondence), such that the Systemengages in Inference carried out by the System maintains the correspondenceThe sterotypical neural pattern induced by a Stimulus is considered a Representation, at least by those who subscribe to Representationalism.
AttentionPerceptionBroad sense: Generative Model that is aware of some Stimulus, reflected by its SalienceThe Model instantly updated to the new Sensory Data because it was paying maximal Attention to the Stimulus.
EvidencePerceptionData as recognized and interpreted by Generative Model of AgentEvery photon is like a piece of Evidence on the retina.
ObservationPerceptionThe Belief updating of an Internal State registered by a Sensory input, given the weighting assigned to that class of input in comparison with weighting of the competing Priors. (This is a narrow sense of “observation,” where the Agent is “looking for this kind of input.” This sense excludes situations where (a) an incoming stimulus with these attributes has already been explained-away or pre-discounted, or (b) the prior is so strongly weighted as to exclude updating in response to any inputs (other than, perhaps, “catastrophic” ones, as may occur in e.g. fainting, hysterical blindness).)One Observation can make all the difference.
PerceptionPerceptionPosterior State Inference after each new Observation.Visual Perception gives us many demonstrations of the characteristics of our Generative Model — for example saccades, the blind spot, and blink supression.
SaliencePerceptionThe extent to which a Cue commands the Attention of an Agent given their Regime of AttentionSalience is related to how relevant a given Stimulus appears to be.
Hierarchical ModelSystemsA hierarchy of Estimators, which operate at different spatiotemporal timescales (so they track features at different scales); all carrying out Predictive ProcessingA Nested Model is a Hierarchical Model, for example the Hierarchically Mechanistic Mind.
Living systemSystemsAgent engaged in AutopoiesisA body is a Living system.
Multi-scale systemSystemsRealism framing of Hierarchical ModelThe Generative Model of counties within states within countries, was a Multi-scale system
SystemSystemsSet of relations described by State space of a Model .George Mobus argues that Systems have some fundamental properties such as structure and function.
Abstract Accuracy
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
abstractCounterpart(abstractCounterpart ?AB ?PHYS) relates a Physical entity to an Abstract one which is an idealized model in some dimension of the Physical entity.
Action and Planning as Divergence Minimization
Action at a distance
Action Integral
Active Blockference
Active learning
active processes
Active Visionrefers to the process of visual perceptions, in terms of oculomotor Sensorimotor Behavior and Cognitive System Generative ModelActive Vision
affect
Agency based model
Agency free model
Algorithm
Alignment
analogy
Analytical Philosophy
anticipation
Appraisal theories of emotion
Artificial
Attenuation of response
Attracting set
Augmented reality
AutopoiesisPhenomena of a System that recapitulates the material and informational causes of its own composition/existence.In the right niche, cells can be considered to exhibit Autopoiesis at the System level.
Bayes-optimal control
Bayesian
Bayesian belief updating
Bayesian Brain
Bayesian mechanics
Bayesian Model Selection
Bayesian surprise
Bethe approximation
Blanket index
Bottom-up attentional control
categorical
category
Category Theory
changing mind (cognition)
changing the mind
changing the world
changing world (action)
chaos
Circular causality
Co-category
Coarse graining
Cognitive Science
Cognitive System
Cognitivism
colimit
Collective behavior
Conceptual metaphor
Conditional density
Conditional Probability
Confidence
Congruence
Connectionism
constraint
Continental Philosophy
Continuous state space
Control (states)
Control theory
Counterfactual
Cybernetics
Decision-makingWithin Active Inference , this is the same as Policy selection
Deflationary
Density
Deontic Action
DeSci
Development
Discrete state space
Dissipation
Distribution
divergence
Domain
Domain-generality
Domain-specificity
Dynamic Causal Modelling
Dynamic expectation maximization
Dynamicism
EcoEvoDevoEcology, Evolution, Development
Ecology
Effective
Effectivity
Embedded Embodied Encultured Enactive Inference
Embodied Belief
Embodied Cybernetic Complexity
Emotion
Empirical prior
Enactivism
Entropy
Epistemic foraging
Ergodicity
Estimator
Evolution
Expectation maximization
Expected Utility Theory
Experience of body ownership
Explaining Away
Explanation
Extended Cognition
Exteroception
Factor graph
Falsification
Far-from-equilibrium
Filter
fitness
flow
Fokker-Planck Equation
Foraging
Forney
frequentist
Friction
Friston's Law
functional magnetic resonance imaging (FMRi)
Functor
Gauge theory
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 Decomposition
Helmholtz Free Energy
Helmholtz machine
Hermeneutics
Hierarchically Mechanistic Mind
High roadOne of two roads (arguments) that lead to the Free Energy Principle as a possible conclusion which starts with philosophical questions about what properties a thing must have to “exist” (i.e. it must be measurable) and then uses principles of Autopoiesis and non-equilibrium Thermodynamic systems from a statistical perspective to show what kinds of systems could continue to maintain themselves over time (see Friston 2019: Beyond the Desert Landscape and the other road, the Low road ). “The high road stands in for a top-down approach that starts by asking fundamental questions about the necessary properties things must possess if they exist. Using mathematical (variational) principles, once can then show that existence is an embodied exchange of a creatures with its environment - that necessarily entails predictive processing as one aspect of self-evidencing mechanics.”The High road to the Free Energy Principle starts by talking about random Non-linear dynamical systems in general without a specific focus on biological organisms with brains.
Homeostasis
Homeostatic system
Hyperprior
Hypothesis
Inflationary
Information bottleneck
Instrumentalism
intelligence
Interface
Interoception
Interoceptive sensitivity
Interpretation
Inverse problem
Kullback-Leibler Divergence
Lagrangian
Latent cause
Lateral geniculate nucleus
Least action
Lens
Likelihood
Low roadOne of two roads (arguments) that lead to the Free Energy Principle as a possible conclusion which starts with fundamental questions from neuroscience and psychology about the nature of perception in biological organisms within a changing environment (see Friston 2019: Beyond the Desert Landscape and the other road, the High road ). “The low road is to pursue the agenda established by Kant and Helmholtz to generalize - in a bottom up way - the capacity for inference and prediction to see how far it takes us in understand embodied exchange with the environment.”The Low road to the Free Energy Principle starts by looking at how biological organisms perceive their environment and take actions within it to develop a notion about how they can successfully predict the next state they will be in (Perception as Hypothesis testing).
Marginal approximation
Markovian MonismMarkovian Monism
Marr's Levels of Description
matching
Material science
maximum caliber
Mean
Mean field approximation
Memory
Message Passing
Mismatch negativity
Mode
Model
Model accuracy
model evidence
Model InversionModel Inversion
Monad
morphism
Morphogenesis
Multisensory integration
Natural
Nested
Network
Neuronal Ensemble
Niche construction
Noisy signal
Non-linear dynamical systems
normative
Optimal control
overfitting
Partially Observed Markov Decision Process
Partition
path
Path integral
Path of Least Action
phenotype
Policy posterior
Policy prior
population
Precariousness
Precision
Prediction error
Prediction error minimization
Predictive Coding
predictive machine
Predictive Processing
PrinciplePrinciple
Probability distribution
Proprioception
Quantum
Quantum mechanics
Quantum-like
Qubit
Random variable
RealismRealism
Receptive field
Recognition density
Renormalization
Representationalism
Reservoir Computing
Reward
RiskRisk
Sample space
Selection bias
Selection history
Self-organization
Selfhood
Semi-Markovian
Sense of agency
Sensorimotor
Sensory attenuation
Sensory Data
Sensory input
Sensory observation
Sensory outcome
Sensory State
Sensory states
Sentiencebeing ‘‘responsive to sensory impressions’’ through adaptive internal processes
Shared Generative Model
Signal
Simulation
solenoidal
Sophisticated Inference
spike-timing dependent plasticity
Statistical manifold
Statistical Parametric Mapping
Stigmergy
Stochastic
Subjective feeling states
sufficient statistic
Surprisal
Swarm
Symbol
Synergetics
T-Maze
Teams
Theory
Thermodynamic system
Thermostatistics
Thing
Thinking Through Other Minds
time
Top-down attentional control
tracking
Umwelt
Unidirectionality
Update
Variance
Variational message passing
Variational Niche Construction
Variational principle
Von Economo neurons
Weak mixing
Working memory
World States
CueInformationa Stimulus , event, object , or Guidance signal that serves to guide Behavior , such as a retrieval cue, or that acts as a @Signal to the presentation of another stimulus, event, or object, such as an unconditioned stimulus or reinforcement. (dictionary.apa.org)
Information GeometryInformationA Statistical manifold each of whose points corresponds to a Probability distribution (e.g. the expectation and variance of a normal density).
Action PredictionActionInference on current and future Expectation of ActionThe Generative Model over the next few timesteps with respect to Active States , is the Action Prediction .
Process TheoryFree EnergyA hypothesis or Model proposal for how a Principle is realized (e.g. the Free Energy Principle)Friston in 2018 wrote “The distinction is between a State [theory] and Process Theory ; i.e., the difference between a normative principle that things may or may not conform to, and a Process Theory or hypothesis about how that principle is realized”
VariationalFree EnergyBiologically: Said of the behavior of a Neuronal Ensemble that minimizes the error of quantities selected by that ensemble, by implementing an algorithm that approximates calculus of variations.
Hidden StateMarkov PartitioningUnobserved variable in Bayesian Inference , can reflect a Latent cause .
CultureAgents in the NicheCulture is the Niche for social Agent, that structures their Regime of Attention
NarrativeAgents in the NicheInformation used by Agent in context of entity- and event-oriented Cognition, specifically Hierarchical Model (Hierarchically Mechanistic Mind ).
NoveltyPerceptionThe Internal State assumed by an Agent‘s epistemic Affordance, when unable to immediately (e.g. locally) resolve Uncertainty about the contingencies — i.e. the opportunity to resolve Uncertainty about ‘what would happen if I did that?’ (The Precision of this assumed Internal State has a distinctive, e.g. multimodal, distribution, i.e. exhibits Ambiguity .)
Regime of AttentionPerceptionfeedback mechanisms among practices in a Culture of scaffolding individuals’ Attention, that guide Agents’ style of Attention; act as determined by bodily, language, and contextual Cues in a given community; and are encoded in higher levels of the cortical hierarchy.
active
area
backbone
brain
causality
classical physics
computer
concentration
concept
condition
consensus
conversation
current
default-mode
dynamics
ecosystem
ego
energy
environment
error
feedback
field
framework
freeFor no cost.
genetic
hierarchical
idea
increase
influence
inverse
language
Logic
machine
matrix
neuronal
object
objective
observer
opportunity
parameter
part
perceptual inference
perspective
phase
physics
play
probability
Probably Approximately Correct
problem
propositional
Propositional attitude
Psychological attitude
purpose
question
random
recognition
relative entropySynonym for Kullback-Leibler Divergence
representsSUMO relation (represents ?THING ?ENTITY) means that ?THING in some way indicates, expresses, connotes, pictures, describes, etc. ?ENTITY. The Predicates containsInformation and realization are subrelations of represents.
resource
role
science
selection
situation
social
states
Stimulus
technology
trajectory
transition
tree
understanding
witness

关键表面

一瞥主动推断本体论

相关项目

共享主题的项目

基于公共项目数据源共享话题和类别计算。

Research Active Inference Journal

A community publication and knowledge channel for Active Inference research and discussion.

Shared topics: Active Inference, Decentralized Science, Eduactive Research Generalized Notation Notation

A text-based notation system for communicating generative models and Active Inference systems that provides a shared language for describing models in a human-r

Shared topics: Active Inference, Knowledge Engineering, Knowledge Representation Research Knowledge Engineering

Knowledge Engineering develops and maintains the public knowledge infrastructure for the Active Inference Institute, including the public frontend, literature m

Shared topics: Active Inference, Knowledge Engineering, Ontology Research Action Research on Collective Foraging

Action Research on Collective Foraging (Negotiation Affordances) applies Active Inference to collective foraging behavior — studying how groups form coalitions,

Shared topics: Active Inference, Eduactive Research Active Inference Cycle Book for Self-Knowing

A reflective book project applying Active Inference as a framework for self-knowledge and personal development.

Shared topics: Active Inference, Eduactive Research Active Inference for Social Sciences

Develops courses, curricula, research, and writing connecting Active Inference to social sciences disciplines.

Shared topics: Active Inference, Eduactive

相关资源

此页的公共链接

外部链接从共享注册表解析，这样面向访问者的目的地保持集中和可检查。

Repository / Projects

GitHub organization

Audience: Developer

Public GitHub organization for Institute repositories and open-source work.

projectsgithub-org
ProjectsPrograms
Repository / Projects

GEO-INFER repository

Audience: Developer

Geospatial modeling repository connected to ecological and bioregional applications.

projectsgeo-infer
ProjectsEcosystem

官方页面

官方机构表面

仓库

相关开源仓库

Repository / Research

act_inf_metaanalysis

Audience: Researcher

Computational meta-analysis of Active Inference literature with nanopublication and knowledge-graph outputs.

TeX / 4 stars / updated 2026-05-04

researchknowledgetex
Learning and ResearchEcosystem
Repository / Research

Active_Inference_Ontology

Audience: Researcher

Ontology-oriented repository for shared Active Inference concepts and decentralized science knowledge infrastructure.

Unspecified / 14 stars / updated 2026-05-18

researchknowledgeunclassified
Learning and ResearchEcosystem
Repository / Projects

ActiveBlockference

Audience: Developer

Notebook-based applied Active Inference work connected to blockchain-adjacent and generative modeling examples.

Jupyter Notebook / 33 stars / updated 2026-05-27

projectrepositoryjupyter-notebook
ProjectsEcosystem
Repository / Projects

ActiveInferAnts

Audience: Developer

Python models and materials for ant-inspired multiagent Active Inference.

Python / 29 stars / updated 2026-05-18

projectrepositorypython
ProjectsEcosystem

查看所有52个公开仓库

相关页面

继续浏览网站

学习者、教育者、课程开发人员、教科书贡献者和对积极推断教育感兴趣的任何人。 教育活跃单元

EduActive 是研究所的教育部门。它负责管理文本书小组、课程、本体工作、积极推断期刊、音频视觉制作、季节性学校以及相关学习计划。EduActive 项目将积极推断与学习基础设施、教学方法和公共教育资源相连接。

 寻求具体工作的建造者、研究人员、维护者、合作伙伴和贡献者。 项目

项目将共享想法转化为仓库、研究成果、学习基础设施、媒体、活动和应用工作。

 知识工程师、信息科学家、研究人员和贡献者，致力于结构化知识资源。 知识工程

知识工程开发和维护着“主动推断研究所”的公共知识基础设施，包括公共前端、文献元分析以及组织知识系统。该项目将所研究成果与更广泛的文献相连接，使其可被机器读取并导航。

 新手、学习者、研究人员、构建者和合作者。 主动推断

主动推断是研究所教育、研究、应用和生态系统支持的核心科学和实践框架。