Bayesian Model-Based Reinforcement Learning

RL variant that maintains beliefs over environment model; compared to active inference using Thompson sampling.

Neighborhood — ranked by edge-count

method

Thompson Sampling
implements
A Bayesian exploration strategy that samples from the posterior distribution over model parameters to decide actions.

concept

Reinforcement learning (RL)
implements
Machine learning paradigm where agents learn to maximize cumulative reward through interaction.

dataset

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

framework

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

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.

Monte-Carlo reinforcement learningmethod0.835
Reinforcement learning methods that update parameters at the end of an episode based on sampled returns.
Bayesian model evidenceconcept0.818
The probability of sensory data under a generative model; negative log evidence is bounded by free energy.
Bayesian Model Selectionconcept0.799
Choosing among candidate models based on model evidence.
There is an implicit behavioral equivalence between Bayesian model-based reinforcement learning and active inference when prior preferences are treated as a reward function.claim0.799
§3, reward shaping conclusion.
Reinforcement Learning from Human Feedbackmethod0.798
Method for fine-tuning LMs based on human preferences; mentioned as combining RL and LMs.
Deep Reinforcement Learningmethod0.795
AI training method inspired by behaviorism, used for autonomous cars and drones; cited as bioinspired success
Reinforcement Learning from AI Feedbackframework0.794
Variant of RLHF where human feedback is replaced with AI-generated feedback for harmlessness.
Bayesian Model Averagingconcept0.791
Predictions formed by averaging over policy-specific beliefs, weighted by policy probabilities.