mlsanityData to 14 Oct 2026

PA2D-MORL: Pareto Ascent Directional Decomposition based Multi-Objective Reinforcement Learning

arXiv 2603.19579 ↗First posted 20 Mar 2026cs.AI cs.LGcited by 2

On the map: Deep reinforcement learning methods → RL for games and combinatorial optimization

Abstract (arXiv)

Multi-objective reinforcement learning (MORL) provides an effective solution for decision-making problems involving conflicting objectives. However, achieving high-quality approximations to the Pareto policy set remains challenging, especially in complex tasks with continuous or high-dimensional state-action space. In this paper, we propose the Pareto Ascent Directional Decomposition based Multi-Objective Reinforcement Learning (PA2D-MORL) method, which constructs an efficient scheme for multi-objective problem decomposition and policy improvement, leading to a superior approximation of Pareto policy set. The proposed method leverages Pareto ascent direction to select the scalarization weights and computes the multi-objective policy gradient, which determines the policy optimization direction and ensures joint improvement on all objectives. Meanwhile, multiple policies are selectively optimized under an evolutionary framework to approximate the Pareto frontier from different directions. Additionally, a Pareto adaptive fine-tuning approach is applied to enhance the density and spread of the Pareto frontier approximation. Experiments on various multi-objective robot control tasks show that the proposed method clearly outperforms the current state-of-the-art algorithm in terms of both quality and stability of the outcomes.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in RL for games and combinatorial optimization, one of 15 directions in Deep reinforcement learning methods. Builds on 7 of its 23 references that are in the map. First posted on arXiv 20 Mar 2026.

Builds on

7 of 23 references.

References · 7

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2016Benchmarking Deep Reinforcement Learning for Continuous Controlcited by 371
  3. 2021A Practical Guide to Multi-Objective Reinforcement Learning and Planningcited by 256
  4. 2019A Generalized Algorithm for Multi-Objective Reinforcement Learning and Policy Adaptationcited by 177
  5. 2016Multi-Objective Deep Reinforcement Learningcited by 83
  6. 2018Meta-Learning for Multi-objective Reinforcement Learningcited by 34
  7. 2021Online certification of preference-based fairness for personalized recommender systemscited by 22

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026ORACLE: A Multi-Objective Reinforcement Learning-Based Analog Circuit Design Optimizer with Large Language Models-Guided Explorationcited by 1
  2. 2026TANGO: RNA Topology and Geometry Co-Designcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Traversing Pareto Optimal Policies: Provably Efficient Multi-Objective Reinforcement Learning
  2. 2025Pareto Set Learning for Multi-Objective Reinforcement Learning
  3. 2024C-MORL: Multi-Objective Reinforcement Learning through Efficient Discovery of Pareto Front
  4. 2024Learning Pareto Set for Multi-Objective Continuous Robot Control
  5. 2025Limitations of Scalarisation in MORL: A Comparative Study in Discrete Environments
  6. 2023Scaling Pareto-Efficient Decision Making Via Offline Multi-Objective RL
  7. 2025Variance Reduced Policy Gradient Method for Multi-Objective Reinforcement Learning
  8. 2025Population-Free Pareto Tracking for Sample-Efficient Multi-Policy MORL