Active inference and agency: optimal control without cost functions

DOI10.1007/s00422-012-0512-8zbMath1267.90167OpenAlexW2091405003WikidataQ39574261 ScholiaQ39574261MaRDI QIDQ353847

Spyridon Samothrakis, Read Montague, Karl J. Friston

Publication date: 16 July 2013

Published in: Biological Cybernetics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1007/s00422-012-0512-8

zbMATH Keywords

optimal controlfree energyinferenceactionBayesianagencypartially observable Markov decision processes

Mathematics Subject Classification ID

Markov and semi-Markov decision processes (90C40)

Related Items (14)

Active inference on discrete state-spaces: a synthesis ⋮ Optimal speech motor control and token-to-token variability: a Bayesian modeling approach ⋮ Robot navigation as hierarchical active inference ⋮ Reward Maximization Through Discrete Active Inference ⋮ Deep active inference as variational policy gradients ⋮ Bayesian optimal control for a non-autonomous stochastic discrete time system ⋮ The Discrete and Continuous Brain: From Decisions to Movement—And Back Again ⋮ Sustained sensorimotor control as intermittent decisions about prediction errors: computational framework and application to ground vehicle steering ⋮ Generalised free energy and active inference ⋮ A tutorial on variational Bayes for latent linear stochastic time-series models ⋮ A Minimum Free Energy Model of Motor Learning ⋮ Modeling the subjective perspective of consciousness and its role in the control of behaviours ⋮ Active Inference: Demystified and Compared ⋮ Predictive Processing in Cognitive Robotics: A Review

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Planning and acting in partially observable stochastic domains
• Action understanding and active inference
• Free energy, value, and attractors
• Dual-control theory. I
• Variational Bayesian identification and prediction of stochastic nonlinear dynamic causal models
• A complete class theorem for statistical problems with finite sample spaces
• Simple statistical gradient-following algorithms for connectionist reinforcement learning
• \({\mathcal Q}\)-learning
• Adaptive dual control. Theory and applications.
• Using Expectation-Maximization for Reinforcement Learning
• Probabilistic Inference and Influence Diagrams
• Proof of the Ergodic Theorem
• On the Theory of Dynamic Programming
• Stochastic Boolean satisfiability

This page was built for publication: Active inference and agency: optimal control without cost functions