Yihang Yao
yihangya[at]andrew.cmu.edu
Hi, welcome to my website! I am a third-year Ph.D. candidate in Safe AI Lab at Carnegie Mellon University, advised by Prof. Ding Zhao. Before CMU, I received my Bachelor's degree from Zhiyuan College, Shanghai Jiao Tong University (SJTU) in 2022. In 2021, I spent a wonderful time working as a research intern in the Intelligent Control Lab led by Prof. Changliu Liu at CMU.
Yihang Yao*, Zhepeng Cen*, Wenhao Ding, Haohong Lin, Shiqi Liu, Wenhao Yu, Tingnan Zhang, Ding Zhao
NeurIPS 2024
TL;DR: We investigate offline RL from a data-centric perspective and propose a diffusion model-based data generator to curate training datasets aligned with user preferences.
Zhepeng Cen, Yihang Yao, Zuxin Liu, Ding Zhao
ICML 2024
TL;DR: We introduce FCSRL, a framework that improves safety constraint estimation in RL through representation learning and self-supervised techniques.
Yihang Yao, Zuxin Liu, Zhepeng Cen, Peide Huang, Tingnan Zhang, Wenhao Yu, Ding Zhao
L4DC 2024
TL;DR: We introduce GradS, a gradient-based method for improving training efficiency in multi-constraint RL by manipulating gradients, optimizing both reward and constraint satisfaction.
Yihang Yao*, Zuxin Liu*, Zhepeng Cen, Jiacheng Zhu, Wenhao Yu, Tingnan Zhang, Ding Zhao
NeurIPS 2023
TL;DR: We introduce CCPO, a framework for versatile/adaptive safe RL that enables efficient training and zero-shot adaptation to varying safety constraints.
Zhepeng Cen, Zuxin Liu, Zitong Wang, Yihang Yao, Henry Lam, Ding Zhao
ICLR 2024
TL;DR: We introduce CDE, a DICE-based method, which addresses OOD errors in offline RL, achieving SOTA results on the D4RL benchmark, particularly in sparse reward and low-data scenarios.
Zuxin Liu*, Zijian Guo*, Haohong Lin, Yihang Yao, Jiacheng Zhu, Zhepeng Cen, Hanjiang Hu, Wenhao Yu, Tingnan Zhang, Jie Tan, Ding Zhao
Journal of Data-centric Machine Learning Research (DMLR); RSS 2023 Safe Autonomy Workshop (Spotlight)
TL;DR: We present a comprehensive benchmarking suite for offline safe RL, featuring expertly crafted safe policies, diverse datasets, and baseline implementations across 38 tasks, designed to accelerate the development and evaluation of safe RL algorithms in both training and deployment phases.
Paper / Website / Code (OSRL) / Code (DSRL) / Code (FSRL)
Zuxin Liu*, Zijian Guo*, Yihang Yao, Zhepeng Cen, Wenhao Yu, Tingnan Zhang, Ding Zhao
ICML 2023
TL;DR: We propose CDT for offline safe RL, which leverages a multi-objective optimization approach to balance safety and task performance, achieving superior adaptability, robustness, and high-reward policies with zero-shot adaptation capabilities.
Zuxin Liu*, Zijian Guo*, Zhepeng Cen, Huan Zhang, Yihang Yao, Hanjiang Hu, Ding Zhao
ICML 2023
TL;DR: We propose SAFER, a robust off-policy learning approach for safe RL that improves policy robustness without adversarial training.
Yihang Yao, Tianhao Wei, Changliu Liu
TL;DR: We present a computationally efficient method for safe control in non-control-affine systems, using energy-function extensions and hyperparameter optimization to ensure safety and efficiency, with theoretical guarantees and numerical validation.