Research map / Reinforcement learning / Policy gradient and actor-critic
True Online Temporal-Difference Learning
Harm van Seijen, A. Rupam Mahmood, Patrick M. Pilarski, Marlos C. Machado, Richard S. Sutton
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 |
|---|---|---|---|
| use true online TD(lambda) instead of accumulate TD(lambda) | TD(lambda) · Random MRPs, three MRPs and three representations | normalized MSE at optimal alpha | Improved |
| use true online TD(lambda) instead of replace TD(lambda) | TD(lambda) · Random MRPs, three MRPs and three representations | normalized MSE at optimal alpha | Improved |
| use true online TD(lambda) instead of accumulate or replace TD(lambda) | TD(lambda) · Myoelectric prosthetic arm (angle and force prediction) | normalized mean absolute return error | Improved |
| use true online Sarsa(lambda) instead of accumulate or replace Sarsa(lambda) | Sarsa(lambda) · ALE Asterix | return per episode | No clear effect |
| use replace traces instead of accumulate traces | TD(lambda) · Myoelectric arm, grip force prediction | normalized return error | Improved |
| use replace traces instead of accumulate traces | TD(lambda) · Myoelectric arm, motor angle prediction | normalized return error | Hurt |
| use replace traces on a single always-active binary feature | TD(lambda) · Two-state example | asymptotic RMS error | Hurt |
| use true online TD(lambda) instead of accumulate TD(lambda) with non-binary features | TD(lambda) · MRP (10, 3, 0.1), non-binary features | normalized MSE | Improved |
Closest papers by meaning
- Temporal Difference Learning: Why It Can Be Fast and How It Will Be Faster (ICLR 2025 · similarity 0.91)
- Towards Parameter-Free Temporal Difference Learning (ICML 2026 · similarity 0.90)
- Discerning Temporal Difference Learning (AAAI 2024 · similarity 0.88)
- An Improved Finite-time Analysis of Temporal Difference Learning with Deep Neural Networks (ICML 2024 · similarity 0.87)
- Almost Sure Convergence of Linear Temporal Difference Learning with Arbitrary Features (NeurIPS 2026 · similarity 0.87)
- The surprising efficiency of temporal difference learning for rare event prediction (NeurIPS 2024 · similarity 0.86)
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.
