Research map / Bandits and online learning / RL theory and MDPs
Reinforcement Learning with History Dependent Dynamic Contexts
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use transformer-based dense history model (Hist-MuZero, stack of 30 observations) instead of MuZero state | MuZero on MovieLens recommendation environment · AttractionEnv / NoveltyEnv, varying alpha | return | Hurt |
| Replace MuZero with DCZero (aggregated-feature confidence augmented state, threshold optimistic value) | MuZero on AttractionEnv and NoveltyEnv · MovieLens movie recommendation environments, alpha=0.99 | return | Helped |
| Increase history discount alpha (longer history dependence) for DCZero vs baselines on AttractionEnv | DCZero · AttractionEnv | return gap over baselines | Helped |
Closest papers by meaning
- MultiScale Contextual Bandits for Long Term Objectives (NeurIPS 2025 · similarity 0.87)
- RL in Latent MDPs is Tractable: Online Guarantees via Off-Policy Evaluation (NeurIPS 2024 · similarity 0.85)
- Bridging State and History Representations: Understanding Self-Predictive RL (ICLR 2024 · similarity 0.84)
- Eluder-based Regret for Stochastic Contextual MDPs (ICML 2024 · similarity 0.84)
- Dynamic Regret of Adversarial MDPs with Unknown Transition and Linear Function Approximation (AAAI 2024 · similarity 0.84)
- Learning to Steer Markovian Agents under Model Uncertainty (ICLR 2025 · similarity 0.84)
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.