Reproduce “Satisficing as competitive Bayesian weighting: a minimal model of goal-directed choice under an unattained target” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Moto Shinriki, Yu Kono, Tatsuji Takahashi
The authors reinterpret RS as ACB, a Bayesian-like estimate that combines an option’s sample mean with an aspiration level, weighted by its share of all trials. Sampling other options moves an abandoned option’s estimate toward aspiration without deleting observations. In simulations, adjusting an unattainable initial target reduces regret; this finding is limited to stationary Bernoulli bandits and does not establish biological behavior.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.