Seven Reward Function Groups

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

• Reward Function Categoriesmethod0.822
  Seven categories determined by which components of f[h] are activated: Objective only, Expect only, Compare only, and combinations
• Reward Functionconcept0.788
  In RL, a scalar signal from the environment that defines the agent's goal; in active inference, reward is just another observation with associated preference.
• Optimal Reward Frameworkframework0.731
  Framework from Singh, Lewis, and Barto 2009 used to select best-performing reward functions via grid search
• Grid search covers 312,130 subjective reward functions per environment after removing duplicatesfinding0.723
  Scale of the hyperparameter search establishing thoroughness of optimization
• Seven frieze pattern groupsframework0.716
  Classification of one-dimensional periodic patterns into 7 symmetry groups using international notation pxyz
• Seven Frieze Symmetry Groupsframework0.711
  Mathematical classification system for one-dimensional periodic patterns using group theory; core framework for the analysis.
• Happiness Function f[h]method0.692
  Subjective reward signal from Dubey et al. 2022 balancing objective reward, expectations, and comparisons; extended in this paper
• Reward Hypothesisconcept0.692
  The claim in RL that any goal can be expressed as maximizing the expected cumulative sum of a scalar reward signal.