mlsanityData to 13 Oct 2026

The Max-Min Formulation of Multi-Objective Reinforcement Learning: From Theory to a Model-Free Algorithm

arXiv 2406.07826 ↗First posted 12 Jun 2024cs.LG cs.AIcited by 14

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

Abstract (arXiv)

In this paper, we consider multi-objective reinforcement learning, which arises in many real-world problems with multiple optimization goals. We approach the problem with a max-min framework focusing on fairness among the multiple goals and develop a relevant theory and a practical model-free algorithm under the max-min framework. The developed theory provides a theoretical advance in multi-objective reinforcement learning, and the proposed algorithm demonstrates a notable performance improvement over existing baseline methods.

Key facts. Cited by 14 papers in the map (data through 13 Oct 2026). Placed in RL for games and combinatorial optimization, one of 15 directions in Deep reinforcement learning methods. Builds on 18 of its 35 references that are in the map. First posted on arXiv 12 Jun 2024.

Builds on

18 of 35 references.

References · 18

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,223
  2. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,656
  3. 2013Playing Atari with Deep Reinforcement Learningcited by 3,221
  4. 2021Convex optimizationcited by 2,488
  5. 2017Reinforcement Learning with Deep Energy-Based Policiescited by 660
  6. 2019Behavior Regularized Offline Reinforcement Learningcited by 657
  7. 2021A Practical Guide to Multi-Objective Reinforcement Learning and Planningcited by 242
  8. 2014A Survey of Multi-Objective Sequential Decision-Makingcited by 232
  9. 2019A Generalized Algorithm for Multi-Objective Reinforcement Learning and Policy Adaptationcited by 168
  10. 2015Taming the Noise in Reinforcement Learning via Soft Updatescited by 125
  11. 2018Dynamic Weights in Multi-Objective Deep Reinforcement Learningcited by 97
  12. 2013Projection onto the probability simplex: An efficient algorithm with a simple proof, and an applicationcited by 65
6 more
  1. 2020Learning Fair Policies in Multiobjective (Deep) Reinforcement Learning with Average and Discounted Rewardscited by 60
  2. 2022PD-MORL: Preference-Driven Multi-Objective Reinforcement Learning Algorithmcited by 52
  3. 2022Welfare and Fairness in Multi-objective Reinforcement Learningcited by 20
  4. 2022Q-Pensieve: Boosting Sample Efficiency of Multi-Objective RL Through Memory Sharing of Q-Snapshotscited by 9
  5. 2023Fairness in Preference-based Reinforcement Learningcited by 9
  6. 2020Quantifying the Impact of Non-Stationarity in Reinforcement Learning-Based Traffic Signal Controlcited by 6

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2025: 3; 2026: 11.

14 papers.

Sort:

By area:

  1. 2026Flow Actor-Critic for Offline Reinforcement Learningcited by 6
  2. 2025Multi-Objective Reinforcement Learning with Max-Min Criterion: A Game-Theoretic Approachcited by 5
  3. 2026Focal Reward: Balanced Reinforcement Learning under Rubric-Based Rewardscited by 4
  4. 2025FairDICE: Fairness-Driven Offline Multi-Objective Reinforcement Learningcited by 2
  5. 2025Reward Dimension Reduction for Scalable Multi-Objective Reinforcement Learningcited by 1
  6. 2026EvoNash-MARL: A Closed-Loop Multi-Agent Reinforcement Learning Framework for Medium-Horizon Equity Allocationcited by 1
  7. 2026Multi-Objective Reinforcement Learning for Large-Scale Tote Allocation in Human-Robot Collaborative Fulfillment Centerscited by 0
  8. 2026[Re] FairDICE: A Fair Tradeoff in Multi-objective Offline RLcited by 0
  9. 2026Auction-Based Online Policy Adaptation for Evolving Objectivescited by 0
  10. 2026MATO: Multi-objective Personalized Alignment with Test-time Optimization for Large Language Modelscited by 0
  11. 2026Constrained Multi-Objective Reinforcement Learning with Max-Min Criterioncited by 0
  12. 2026AETDICE: Unified Framework and Offline Optimization for Nonlinear Multi-Objective RLcited by 0
2 more
  1. 2026RLVR2: Reinforcement Learning with Verifiable Rubric-based Rankingcited by 0
  2. 2026Which Preferences to Train On? End-to-End Multi-Objective Alignment with an Adversarial Preference Distributioncited by 0