Park et al. 2023: Generative agents — interactive simulacra of human behavior

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

• What is the generative model that best explains observable data from a behaving agent?question0.778
  Central challenge for active inference stated in Discussion.
• Simulations of a generative process (done in the laboratory) are essential to test that the defined process generates coherent complex variable morphology before applying it to real landclaim0.777
  Alexander's methodological justification for using the Guasare simulation studies.
• Generative models are entailed by adaptive behavior, not explicitly encoded in brain statesclaim0.776
  Distinction from Bayesian brain: generative model is consequence of dynamics, not neural representation
• What we call GPT's 'downstream behavior' is the behavior of simulacra; it is primarily through simulacra that GPT has potential to perform meaningful work.claim0.774
  Clarifies where agency resides.
• GPT, insofar as it is inner-aligned, is a simulator which can simulate agentic and non-agentic simulacra.claim0.771
  Central thesis of the post.
• Generative models function as control systems that guide adaptive action policy selectionclaim0.766
  Core claim: generative models regulate organism behavior to maintain phenotypic bounds, not represent external world
• Agent Based Model Of Morphogenesisframework0.760
  Computational simulation modeling cells as agents with homeostatic loops interacting through stress sharing.
• Virtual rodent agent (Merel et al. 2019)concept0.760
  RL-trained LSTM agent controlling a realistic 38-DoF rat body in a 3D environment.