AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Modeling Human Exploration Through Resource-Rational Reinforcement Learning

Marcel Binz, Eric Schulz

NeurIPS 2022 · Reinforcement learning › Visual and representation RL · approach: Task-inference meta-RL

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use RR-RL2 instead of the hybrid probit exploration model to fit human choicesRR-RL2 · Two-armed bandit human dataBIC: 6158.91 -> 5562.63Improved
Increase description length of RR-RL2 from 1 to about 100 natsRR-RL2 · Two-armed bandit probit analysisexploration style: average regression coefficient of 0.15Mixed
Increase description length from 100 to 1000 natsRR-RL2 · Two-armed bandit probit analysisThompson sampling weight: average regression coefficient of 0.19Improved
Increase description length to 1000 nats and aboveRR-RL2 · Two-armed bandit probit analysisUCB-based weight: average regression coefficient of 0.23Mixed
Reduce description length of RR-RL2RR-RL2 · Iowa Gambling Taskhigh-risk arm selectionHurt
Increase description length of RR-RL2RR-RL2 · Horizon taskstrategic directed explorationImproved
Increase description length of RR-RL2RR-RL2 · Horizon taskstrategic random explorationNo clear effect
Increase reward noise standard deviation by 50% when retraining (ablation)RR-RL2 · Horizon taskinteraction effect on directed explorationImproved
Increase description length of RR-RL2 (regret)RR-RL2 · Horizon taskregretImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy