基本信息
views: 3
![](https://originalfileserver.aminer.cn/sys/aminer/icon/show-trajectory.png)
Bio
I'm broadly interested in reinforcement learning, large language models, and machine learning. Currently, my research aims to i) understand the structural information of deep RL & LLMs and how to leverage it to improve agent performance in the wild (e.g., dealing with biased, noisy, or redundant data, or extrapolating to unseen tasks/environments), ii) develop controllable AI in both training and inference/adaptation; and iii) theory and real-world application of Human-AI alignment. And Yes we are developing these methods for RL and LLMs.
Our research is built upon empirical and theoretical analysis of the learning dynamics, utilizing tools from stochastic processes, functional analysis, algebra, optimization, information theory, and large language models. Our goal is to develop efficient, stable, trustworthy agents based on coevolution between humans and agents.
Our research is built upon empirical and theoretical analysis of the learning dynamics, utilizing tools from stochastic processes, functional analysis, algebra, optimization, information theory, and large language models. Our goal is to develop efficient, stable, trustworthy agents based on coevolution between humans and agents.
Research Interests
Papers共 14 篇Author StatisticsCo-AuthorSimilar Experts
By YearBy Citation主题筛选期刊级别筛选合作者筛选合作机构筛选
时间
引用量
主题
期刊级别
合作者
合作机构
ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)pp.6970-6974, (2024)
ICLR 2024 (2024)
Cited0Views0EIBibtex
0
0
CoRR (2023): 2454-2456
NeurIPS (2023)
MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES: RESEARCH TRACK, ECML PKDD 2023, PT IV (2023): 573-589
CVPR 2023 (2023): 20215-20225
CoRR (2022)
ArXiv (2022)
arXiv (Cornell University) (2021)
Load More
Author Statistics
Co-Author
Co-Institution
D-Core
- 合作者
- 学生
- 导师
Data Disclaimer
The page data are from open Internet sources, cooperative publishers and automatic analysis results through AI technology. We do not make any commitments and guarantees for the validity, accuracy, correctness, reliability, completeness and timeliness of the page data. If you have any questions, please contact us by email: report@aminer.cn