Research map / Reinforcement learning / Visual and representation RL
Staying up to Date with Online Content Changes Using Reinforcement Learning for Scheduling
Andrey Kolobov, Yuval Peres, Cheng Lu, Eric J Horvitz
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Optimize harmonic staleness cost (LambdaCrawl) instead of binary cost (BLC) | LambdaCrawl · 18.5M URL crawl | harmonic policy cost J_h: 35% | Improved |
| Optimize harmonic cost; evaluate on binary cost | LambdaCrawl · 18.5M URL crawl | binary policy cost J_b | Improved |
| Exploit complete change observations (LC-CO vs LC-IO) (ablation) | LambdaCrawl on complete-observation URLs · Complete-observation URL subset | harmonic policy cost J_h: 2× | Improved |
| Exploit complete change observations (LC-CO vs others) (ablation) | LambdaCrawl on complete-observation URLs · Complete-observation URL subset | harmonic and binary policy cost: 50% | Improved |
| LC-IO (ignore complete observations) compared with BLC and BLC-eps (ablation) | LambdaCrawl · Complete-observation URL subset | binary policy cost J_b | No clear effect |
| Use Prop. 9 approximation (LLCA) instead of exact LLC when rates unknown (ablation) | RL-based LambdaCrawl · Simulated 21-day runs, 100,000-URL subsamples | convergence speed vs asymptotic harmonic cost | Mixed |
| Use approximation (LCA) on binary cost (ablation) | LambdaCrawl · 18.5M URL crawl | binary policy cost J_b | No clear effect |
| RL-based LambdaCrawl (LLC) vs RL-based binary crawler (BLLC-eps) | LLC · Simulated 21-day runs | convergence speed | Improved |
Closest papers by meaning
- Pausing Policy Learning in Non-stationary Reinforcement Learning (ICML 2024 · similarity 0.83)
- Online Reinforcement Learning in Non-Stationary Context-Driven Environments (ICLR 2025 · similarity 0.81)
- DeepDiver: Adaptive Web-Search Intensity Scaling via Reinforcement Learning (NeurIPS 2025 · similarity 0.81)
- From Static Policies to Adaptive Priors in Offline Reinforcement Learning (NeurIPS 2026 · similarity 0.81)
- Simple Ingredients for Offline Reinforcement Learning (ICML 2024 · similarity 0.81)
- A Perspective of Q-value Estimation on Offline-to-Online Reinforcement Learning (AAAI 2024 · similarity 0.81)
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.
