Research map / Reinforcement learning / Policy gradient and actor-critic

Efficient Discovery of Pareto Front for Multi-Objective Reinforcement Learning

ICLR 2025 · Reinforcement learning › Policy gradient and actor-critic · approach: Tree search and policy optimization

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes an efficient method for discovering the Pareto front in multi-objective reinforcement learning, addressing the limitations of fixed policy sets or preference-conditioned policies trained over sampled preference vectors.

Problem

Multi-objective reinforcement learning must handle changing preferences, including unseen ones, but prior dominating MORL methods generate fixed policy sets or preference-conditioned policies only for sampled preference vectors and cannot ensure efficient discovery of the Pareto front.

Approach

Abstract does not specify the method; the stated motivation is moving beyond per-sampled-preference training iterations toward efficient Pareto front discovery.

Closest papers by meaning

• Pareto Set Learning for Multi-Objective Reinforcement Learning (AAAI 2025 · similarity 0.94)
• Preference Conditioned Multi-Objective Reinforcement Learning: Decomposed, Diversity-Driven Policy Optimization (NeurIPS 2026 · similarity 0.93)
• On Generalization Across Environments In Multi-Objective Reinforcement Learning (ICLR 2025 · similarity 0.91)
• PA2D-MORL: Pareto Ascent Directional Decomposition Based Multi-Objective Reinforcement Learning (AAAI 2024 · similarity 0.91)
• An Analytical Study of Utility Functions in Multi-Objective Reinforcement Learning (NeurIPS 2024 · similarity 0.91)
• Preference Controllable Reinforcement Learning with Advanced Multi-Objective Optimization (ICML 2025 · similarity 0.91)