AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Interpretable and Personalized Apprenticeship Scheduling: Learning Interpretable Scheduling Policies from Heterogeneous User Demonstrations

Rohan Paleja, Andrew Silva, Letian Chen, Matthew Gombolay

NeurIPS 2020 · Learning theory and algorithms › Combinatorial and search optimization · approach: Structure and complexity of constraint reasoning · 5 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
discretize PNT into interpretable formPNT · Synthetic schedulingaccuracy: 96.13 -> 99.66Improved
discretize PNT into interpretable formPNT · Synthetic low-dimaccuracy: 97.30 -> 96.13Hurt
discretize PNT into interpretable formPNT · Taxi domainaccuracy: 88.22 -> 77.73Hurt
PNT-learned embeddings appended to states for DT (ablation)Decision tree · Taxi domainaccuracy: 74.90 -> 87.85Improved
DT with embeddings omega instead of vanilla DT (ablation)Decision tree · Synthetic schedulingaccuracy: 32.4 -> 32.85No clear effect
DT with embeddings omega instead of vanilla DT (ablation)Decision tree · Synthetic low-dimaccuracy: 55.76 -> 53.66No clear effect
personalized embedding PNT versus best baseline DDTPNT · Synthetic schedulingaccuracy: 52.35 -> 96.13Improved
PNT versus InfoGAILPNT · Synthetic schedulingaccuracy: 25.72 -> 96.13Improved
pretrain PNT then DAgger instead of random init DAggerPNT · Synthetic schedulingsuccessful schedule completion: +28.57%Improved
pretrain PNT then DAgger instead of random init DAggerPNT · Synthetic schedulingtasks completed before failure: +12.5%Improved
tree-based PNT versus neural networkCounterfactual PNT · User study (20 participants)simulation correctnessImproved
tree-based PNT versus neural networkCounterfactual PNT · User study (20 participants)time to compute outputImproved

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