AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Self-Correcting Models for Model-Based Reinforcement Learning

Erik Talvitie

AAAI 2017 · Reinforcement learning › Visual and representation RL · approach: Inferring decision-making from behavior · 47 citations (OpenAlex)

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 hallucinated training instead of standard DAgger-MC trainingMBRL with a flawed model on Shooter with moving bullseyes · Shooter, moving bullseyesdiscounted returnImproved
Use hallucinated training instead of standard DAgger-MC trainingMBRL when the model class contains a perfect model · Shooter, fixed bullseyesdiscounted returnHurt
Use hallucinated one-step error as a training objective in stochastic environmentsModel-based reinforcement learning · Stochastic MDPsvalidity as a proxy for multi-step prediction errorHurt
Use hallucinated one-step error with a perfect model and a stochastic policyModel-based reinforcement learning · Deterministic MDPs with arbitrary policieshallucinated one-step errorHurt
Use separate models for each rollout depth instead of one shared model (ablation)H-DAgger-MC · Shooter, fixed bullseyesdiscounted returnImproved
Use a single model across rollout depths instead of separate depth-specific models (ablation)H-DAgger-MC · Shooter, fixed bullseyes, training rollouts to depth 15discounted returnHurt
Truncate training rollouts at shorter depths (ablation)H-DAgger-MC with a single shared model · Shooter, fixed bullseyesdiscounted returnImproved
Train on deep hallucinated rollouts from the beginning rather than gradually increasing rollout depth (ablation)H-DAgger-MC · Shootertraining effectivenessHurt

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