Title | ||
---|---|---|
Protecting Reward Function of Reinforcement Learning via Minimal and Non-catastrophic Adversarial Trajectory |
Abstract | ||
---|---|---|
Reward functions are critical hyperparameters with commercial values for individual or distributed reinforcement learning (RL), as slightly different reward functions result in significantly different performance. However, existing inverse reinforcement learning (IRL) methods can be utilized to approximate reward functions just based on collected expert trajectories through observing. Thus, in the... |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/SRDS53918.2021.00037 | 2021 40th International Symposium on Reliable Distributed Systems (SRDS) |
Keywords | DocType | ISSN |
Measurement,Costs,Perturbation methods,Clustering algorithms,Reinforcement learning,Predictive models,Prediction algorithms | Conference | 1060-9857 |
ISBN | Citations | PageRank |
978-1-6654-3819-3 | 0 | 0.34 |
References | Authors | |
0 | 9 |
Name | Order | Citations | PageRank |
---|---|---|---|
Tong Chen | 1 | 0 | 0.34 |
Yingxiao Xiang | 2 | 3 | 3.87 |
Yike Li | 3 | 0 | 1.35 |
Yunzhe Tian | 4 | 1 | 3.05 |
Endong Tong | 5 | 5 | 6.96 |
Wenjia Niu | 6 | 178 | 30.33 |
Jiqiang Liu | 7 | 315 | 52.31 |
Gang Li | 8 | 381 | 62.77 |
Qi Chen | 9 | 261 | 24.99 |