Bayesian inference with limited resources

How to do Bayesian inference with bounded computational resources?

  1. Limited resources: finite automata, polynomial time, etc.
  2. Fundamental limits: undecidability

A possible application that interests me is Bayesian inference in automated theorem proving or computer algebra.

Literature

  • Wilson, 2014: Bounded information and biases in information processing (doi)
    • Original manuscript in 2002 (pdf)
  • Halpern, Pass, Seeman, 2014: Decision theory with resource-bounded agents (arxiv, doi)
    • Review of recent work, emphasizing previous papers by authors
    • Halpern & Pass, 2010: I don’t want to think about it now: Decision theory with costly computation (arxiv, AAAI )
    • Halpern, Pass, Seeman, 2012: I’m doing as well as I can: Modeling people as rational finite automata (AAAI )
  • Ackerman, Freer, Roy, 2011: On the computability of conditional probability (arxiv)