Optimization-as-Inference View of ICL (von Oswald et al. 2023)

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

• Meta-Learning View of ICL (Dai et al. 2023)framework0.826
  Views ICL as a form of meta-learning; UCCT sits alongside this account
• Bayesian ICL (Xie et al. 2022)framework0.793
  Prior framework explaining ICL as inference over task structure; UCCT adopts and extends the Bayesian lens
• Competition Dynamics ICL Framework (Park et al. 2024)framework0.743
  Formalizes regime shifts between retrieval-like and inference-like ICL; UCCT complements with when-predictor
• Active inference treats exploration and exploitation as two different ways of tackling the same problem: minimizing uncertainty.claim0.738
  §2, comment on expected free energy decomposition.
• Constraints and biases of human intelligence are best avoided by ‘designing for emergence’ rather than implementing a pre-existing theory of intelligence.claim0.737
  Recommendation for creating non-anthropocentric machine intelligence.
• Friston, FitzGerald et al. (2016) — Active inference and learningconcept0.736
  Prior active inference paper providing detailed neurophysiological implementation of belief updates
• "active inference separates the problems of optimizing action and perception by assuming that action fulfills predictions based on inferred states"quote0.735
• Active Inference, Curiosity and Insight (Friston et al., 2017)concept0.731
  The primary source paper being extracted