Deep Reinforcement Learning
Deep reinforcement learning (DRL) aims to train agents to make optimal decisions in complex environments by learning through trial and error. Current research focuses on improving DRL's robustness, sample efficiency, and interpretability, often employing architectures like Proximal Policy Optimization (PPO), deep Q-networks (DQNs), and graph neural networks (GNNs) to address challenges in diverse applications such as robotics, game playing, and resource management. The resulting advancements have significant implications for various fields, enabling the development of more adaptable and efficient autonomous systems across numerous domains.
Papers
Improving Environment Robustness of Deep Reinforcement Learning Approaches for Autonomous Racing Using Bayesian Optimization-based Curriculum Learning
Rohan Banerjee, Prishita Ray, Mark Campbell
Analyzing Generalization in Policy Networks: A Case Study with the Double-Integrator System
Ruining Zhang, Haoran Han, Maolong Lv, Qisong Yang, Jian Cheng
MPC-Inspired Reinforcement Learning for Verifiable Model-Free Control
Yiwen Lu, Zishuo Li, Yihan Zhou, Na Li, Yilin Mo
Robotic Control of the Deformation of Soft Linear Objects Using Deep Reinforcement Learning
Mélodie Hani Daniel Zakaria, Miguel Aranda, Laurent Lequièvre, Sébastien Lengagne, Juan Antonio Corrales Ramón, Youcef Mezouar
Canaries and Whistles: Resilient Drone Communication Networks with (or without) Deep Reinforcement Learning
Chris Hicks, Vasilios Mavroudis, Myles Foley, Thomas Davies, Kate Highnam, Tim Watson
Deep Reinforcement Learning for Community Battery Scheduling under Uncertainties of Load, PV Generation, and Energy Prices
Jiarong Fan, Hao Wang
Autonomous and Adaptive Role Selection for Multi-robot Collaborative Area Search Based on Deep Reinforcement Learning
Lina Zhu, Jiyu Cheng, Hao Zhang, Zhichao Cui, Wei Zhang, Yuehu Liu