Targeted adversarial attacks against neural network trajectory predictors

K Tan, J Wang, Y Kantaros - Learning for Dynamics and …, 2023 - proceedings.mlr.press
Trajectory prediction is an integral component of modern autonomous systems as it allows
for envisioning future intentions of nearby moving agents. Due to the lack of other agents' …

Connected superlevel set in (deep) reinforcement learning and its application to minimax theorems

S Zeng, T Doan, J Romberg - Advances in Neural …, 2023 - proceedings.neurips.cc
The aim of this paper is to improve the understanding of the optimization landscape for
policy optimization problems in reinforcement learning. Specifically, we show that the …