community
active
leiden_hybrid_concepts
label: sonnet
community:leiden_hybrid_concepts-run2-c20Active inference & free energy minimization
Friston's framework unifying perception, action, and learning under variational free energy minimization.
14 members. Each node is clickable.
Loading graph…
Drawn from 5 sources
The papers/notes whose extracted claims & findings make up this cluster.
- Active inference: demystified and compared4 members
- Active inference on discrete state-spaces: a synthesis3 members
- Self-Improvising Memory: A Perspective on Memories as Agential, Dynamically Reinterpreting Cognitive Glue3 members
- Active Inference: A Process Theory3 members
- A Free energy principle for the brain (lecture summary)2 members
Bridges (11)
Other communities that share members with this one — cross-cutting threads or papers that sit at the seam between two themes.
- Active inference & agent ecology11 shared
- Active inference and free energy minimization3 shared
- Active inference as Bayes-optimal exploration3 shared
- Distributed biological agency and morphogenetic learning1 shared
- Causal emergence in biological systems1 shared
- Bioelectric memory & morphogenetic identity1 shared
- Empirical operationalization of agency attribution1 shared
- Free energy minimization in active inference1 shared
- Bioelectric computation and morphological intelligence1 shared
- Memory persistence through radical embodied transformation1 shared
- Bioelectric morphogenesis & anatomical intelligence1 shared
Claims (12)
- Acting to maximize value is the same as acting to minimize surprise; value is simply the probability of sensory input expected by an agent.Reinterprets classical reward/value concepts through free energy lens.
- Acting to optimize value and perception are two aspects of exactly the same principle: minimization of free energy.Foundational claim unifying action and perception within single optimization framework.
- Active inference agents can learn their own reward function (prior preferences) by interacting with the environment, bypassing the need for an explicit reward signal.Abstract and §3, preference learning section.
- Active inference describes the dynamics of systems that persist at non-equilibrium steady-state and that can be statistically segregated from their environment via a Markov blanket.Sets the theoretical grounding in Section 2.
- Agents perceive by minimizing variational free energy to ensure model consistency with past observations and act by minimizing expected free energy to make future sensations consistent with preferences.Formalization of perception-action cycle integrating inference and decision-making.
- All neuronal processing and action selection minimize variational free energy, unifying perception, action, and learning.Fundamental assertion: single imperative (free energy minimization) explains diverse cognitive and neural phenomena.
- Behavior prescribed by active inference dynamics is approximately Bayes-optimal.Process theory outcomes produce normatively sound decision-making.
- Biology commits to optimizing salience and meaning, not fidelity, via high-level, agent-based interpretation.Central claim that biological memory prioritizes gestalt over details.
- Expected free energy decomposes into risk (exploitation) and ambiguity (exploration) terms, providing optimal balance between goal-seeking and novelty-seeking.Key insight into structure of decision-making; explains intrinsic motivation and curiosity.
- Natural exploration-exploitation trade-offs emerge automatically from expected free energy minimization without hyperparameter tuning.Active inference achieves Bayes-optimal arbitration between exploration and exploitation without handcrafted mechanisms like ε-greedy.
- Neuronal responses can be described as gradient descent on variational free energy.Central claim: gradient descent on free energy is a valid process-level description of neural activity.
- Salience preservation over fidelity in memory
Findings (2)
- Active inference agents engage in information-seeking behavior in reward-free FrozenLake environments, contrasting with Q-learning but similar to Bayesian RL.Empirical demonstration on FrozenLake; shows epistemic value drives exploration absent reward signal.
- All three agent types (active inference, Q-learning, Bayesian RL) perform adequately in stationary FrozenLake; only active inference achieves Bayes-optimal behavior in non-stationary settings.Key empirical result validating online planning capability of active inference.