8.9 Advancing Investment Frontiers: Industry-grade Deep Reinforcement Learning for Portfolio Optimization
- Authors: Philip Ndikum, Serge Ndikum
- Reason: The paper introduces an innovative industry-grade framework and presents a proprietary RL agent with potential scalable applications in quantitative finance, enhancing its influence by bridging theoretical advancements with real-world applicability.
8.7 Corruption-Robust Offline Two-Player Zero-Sum Markov Games
- Authors: Andi Nika, Debmalya Mandal, Adish Singla, Goran Radanović
- Reason: This paper tackles the practical and challenging problem of learning in the presence of data corruption, presenting robust algorithms and performance guarantees, which is critical for trustworthy reinforcement learning applications.
8.5 The Fusion of Deep Reinforcement Learning and Edge Computing for Real-time Monitoring and Control Optimization in IoT Environments
- Authors: Jingyu Xu, Weixiang Wan, Linying Pan, Wenjian Sun, Yuxiang Liu
- Reason: Combining DRL with edge computing for IoT proposes a novel approach that could yield substantial performance improvements and cost savings in real-time, offering significant potential for influencing the field.
8.3 Do Agents Dream of Electric Sheep?: Improving Generalization in Reinforcement Learning through Generative Learning
- Authors: Giorgio Franceschelli, Mirco Musolesi
- Reason: The paper presents a creative and novel approach to enhance generalization in reinforcement learning agents, inspired by the Overfitted Brain hypothesis and using generative learning techniques.
8.1 $\widetilde{O}(T^{-1})$ Convergence to (Coarse) Correlated Equilibria in Full-Information General-Sum Markov Games
- Authors: Weichao Mao, Haoran Qiu, Chen Wang, Hubertus Franke, Zbigniew Kalbarczyk, Tamer Başar
- Reason: The work contributes to the theoretical understanding of no-regret learning in Markov games, a foundational aspect of multi-agent reinforcement learning, which is important for advancing the capabilities of RL systems.