Research map / Reinforcement learning / Policy gradient and actor-critic
Risk-Averse Learning by Temporal Difference Methods with Markov Risk Measures
Umit Köse, Andrzej Ruszczyński
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 risk-averse TD(lambda) (mean-semideviation) instead of risk-neutral TD(lambda) | TD(lambda) for fleet-assignment MDP · Fleet management, 20,000 stages, alpha 0.95/0.8/0.6, lambda 0/0.5/0.9 | average profit per stage | Mixed |
| use a large lambda (0.9) (ablation) | optimistic risk-averse TD(lambda) · Fleet management, 20,000 stages | average profit per stage | Hurt |
| use risk-averse instead of risk-neutral TD at early training (t=200) | TD(lambda) · Fleet management, 207 trajectories, alpha 0.95 | empirical distribution of profit per stage at t=200 | Improved |
| use risk-averse instead of risk-neutral linear policy evaluation | linear TD policy evaluation · Small fleet (K=12, M=3, 91 states), myopic policy | mean-square error vs full evaluation: risk-neutral final error twice as large as risk-averse | Improved |
| use the policy from risk-averse linear value approximation instead of the myopic policy | myopic policy · Small fleet, one policy improvement step | average profit | Improved |
| use the policy from risk-neutral linear value approximation instead of the myopic policy | myopic policy · Small fleet, one policy improvement step | average profit | Hurt |
Closest papers by meaning
- Risk-Aware Reinforcement Learning with Coherent Risk Measures and Non-linear Function Approximation (ICLR 2023 · similarity 0.89)
- On Dynamic Programming Decompositions of Static Risk Measures in Markov Decision Processes (NeurIPS 2023 · similarity 0.88)
- Risk-averse Total-reward MDPs with ERM and EVaR (AAAI 2025 · similarity 0.87)
- Risk-Averse Total-Reward Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- Risk-Conditioned Reinforcement Learning: A Generalized Approach for Adapting to Varying Risk Measures (AAAI 2024 · similarity 0.86)
- Planning and Learning in Average Risk-aware MDPs (NeurIPS 2025 · 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.
