Raileanu et al. 2018 - Modeling Others Using Oneself in Multi-Agent Reinforcement Learning

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

• Reinforcement learning is sufficient for agency.claim0.797
  Argument that RL meets the agency indicator.
• Reinforcement learning acting on individual characteristics affecting their connections to others can result in dynamics that are equivalent to unsupervised learning at the system scale.claim0.792
  Key insight linking individual rewards to system-level learning.
• Reinforcement Learningframework0.788
  Alternative framework for agent behavior; based on reward maximization rather than free energy minimization.
• Monte-Carlo reinforcement learningmethod0.788
  Reinforcement learning methods that update parameters at the end of an episode based on sampled returns.
• Reinforcement Learning from Human Feedbackmethod0.787
  Method for fine-tuning LMs based on human preferences; mentioned as combining RL and LMs.
• Bayesian Model-Based Reinforcement Learningframework0.785
  RL variant that maintains beliefs over environment model; compared to active inference using Thompson sampling.
• Ouyang et al. 2022: Training language models to follow instructions with human feedbackconcept0.782
  RLHF paper cited as a major fine-tuning technique used in commercial dialogue agents
• Reinforcement learning can be regarded as a limiting or special case of model-based approaches in general — or active inference in particular — when epistemic value is removed.claim0.777
  §3 Discussion.