Jev Bayes, No? Testing TypeSafe AI's Jev against Bayesian-optimal strategies, and testing if Jev can effectivly use Bayesian priors.
Stars measure the whole repository, including work unrelated to Jev.
Documented records the linked documentation or source. JevHunt has not independently run or benchmarked this project.
Ten experiments and a batching validation are complete for `jev-1.13.0` on stationary Bernoulli bandits: **598,080 evaluation/control questions across fixed-state probes and 1,560 independent online environments**. Start with the [concise research summary] for the questions, experiments, results, and interpretation; [conversion instructions] support Word and PDF output. The [research report] provides effect sizes, uncertainty intervals, figures, and interpretation; the [data audit] verifies comp
Evidence commit: e1514233f8d99f53bd49ea971a15078956496f82
Discovered through: github-search.
How we review →