Active inference agent with learnable preferences developed a strict preference for goals (score +) when the Frisbee location was encountered first, becoming a goal-seeking agent.

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

• Active inference agents can learn epistemic habits in the absence of extrinsic rewards through ambiguity minimization.claim0.824
  §2, summarizing information-seeking behavior.
• Active inference agents engage in information-seeking behavior in reward-free FrozenLake environments, contrasting with Q-learning but similar to Bayesian RL.finding0.818
  Empirical demonstration on FrozenLake; shows epistemic value drives exploration absent reward signal.
• Can active inference agents learn their own prior preferences without explicit reward signals?question0.815
  Question answered by the preference learning experiments.
• Under active inference, the ultimate ‘goal’ is to maintain a coherent phenotype and persist over time, not to maximize reward.claim0.815
  §3, preference learning discussion.
• Friston, FitzGerald et al. (2016) — Active inference and learningconcept0.810
  Prior active inference paper providing detailed neurophysiological implementation of belief updates
• Active inference agents can carry out epistemic exploration and account for uncertainty about their environment in a Bayes-optimal fashion.claim0.807
  Abstract and §1, summarizing a key property.
• In active inference, reward can simply be treated as another observation we have a preference over, rather than a special signal.claim0.806
  Abstract; central distinction.
• Under reward shaping (G=100, H=-100, F=0), Active Inference scored 99.52, Bayesian RL 99.77, Q-learning 95.56, with nearly identical behavior between belief-based agents.finding0.803
  Table 2, row 3, showing equivalence when prior preferences match rewards.