1. 9.2 Anytime-Competitive Reinforcement Learning with Policy Prior
   • Authors: Jianyi Yang, Pengfei Li, Tongxin Li, Adam Wierman, Shaolei Ren
   • Reason: Accepted at NeurIPS 2023, the paper addresses an important problem - Anytime-Competitive Markov Decision Process (A-CMDP) - and proposes Anytime-Competitive Reinforcement Learning (ACRL) which shows solid experimental results.
2. 8.8 RiskQ: Risk-sensitive Multi-Agent Reinforcement Learning Value Factorization
   • Authors: Siqi Shen, Chennan Ma, Chao Li, Weiquan Liu, Yongquan Fu, Songzhu Mei, Xinwang Liu, Cheng Wang
   • Reason: NeurIPS 2023 submission, presents a novel method - RiskQ - that models the joint return distribution and showcases promising performance via extensive experiments.
3. 8.5 Robust Adversarial Reinforcement Learning via Bounded Rationality Curricula
   • Authors: Aryaman Reddi, Maximilian Tölle, Jan Peters, Georgia Chalvatzaki, Carlo D’Eramo
   • Reason: Under review, proposes an innovative approach for adversarial RL based on entropy regularization to solve complex saddle point optimization problems, shows strong experimental performance against several benchmarks.
4. 8.3 Score Models for Offline Goal-Conditioned Reinforcement Learning
   • Authors: Harshit Sikchi, Rohan Chitnis, Ahmed Touati, Alborz Geramifard, Amy Zhang, Scott Niekum
   • Reason: Preprint, proposes SMORe - a novel, principled approach to Offline GCRL, capable of leveraging suboptimal offline data, outperforms state-of-the-art baselines in several experimental benchmarks.
5. 8.1 Optimistic Multi-Agent Policy Gradient for Cooperative Tasks
   • Authors: Wenshuai Zhao, Yi Zhao, Zhiyuan Li, Juho Kannala, Joni Pajarinen
   • Reason: Comprehensive 16-pages paper, addresses an essential problem - Relative overgeneralization (RO) - in cooperative multi-agent learning tasks, introduces a novel framework that shows extensive positive results on diverse sets of tasks, including complex Multi-agent MuJoCo and Overcooked benchmarks.