Q-learning

Model-free RL algorithm used in experimental comparison; employs ε-greedy exploration.

Neighborhood — ranked by edge-count

framework

Reinforcement Learning
cites
Alternative framework for agent behavior; based on reward maximization rather than free energy minimization.

concept

Reinforcement learning (RL)
implements
Machine learning paradigm where agents learn to maximize cumulative reward through interaction.
Watkins and Dayan 1992
cites
Original Q-Learning paper cited for the learning algorithm used in all agents

dataset

FrozenLake OpenAI Gym Environment
cites
Modified discrete state-space environment used for experimental comparison of active inference and RL agents.

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

Learningconcept0.839
Inference of parameters encoding contingencies of the world (e.g., likelihood matrix A) at slower timescale than perception.
Physical learningframework0.795
Framework for solving inverse problems in which physical systems autonomously adapt their parameters in response to stimuli through local learning rules, without requiring computational design or explicit cost functions
Perceptual Learningconcept0.795
Process of inferring causes of sensory information; unified with value learning as integral aspects of free energy minimization.
Early learningconcept0.789
The primary domain in which Nicholson's Theory of Loose Parts became influential and known.
Meta-learningconcept0.787
The capability of GPT-3 to learn tasks from few-shot prompts during runtime.
Epistemic Learningconcept0.786
Learning model parameters through curious, uncertainty-reducing behavior; reducing ignorance about contingencies
Associative Learningconcept0.786
Sentience criterion; capacity occurs even in gene regulatory networks and non-neural morphogenetic agents.
Concept Learningconcept0.784
Acquisition of new concepts by Bayesian model expansion and reduction.