Gradient-Adaptive Pareto Optimization for Constrained Reinforcement Learning
Zixian Zhou, Mengda Huang, Feiyang Pan, Jia He, Xiang Ao, Dandan Tu, Qing He.To appear in AAAI 2023.
Feiyang Pan Homepage
About me
I received my PhD degree in June 2021, under the supervision of Prof. Qing He at Machine Learning and Data Mining group, Institute of Computing Technology, Chinese Academy of Sciences (ICT, CAS). During my PhD, I won the XiaPeisu Scholarship, the top award of ICT, CAS. I was also a PhD visitor at IIIS, Tsinghua University, under supervision of Prof. Pingzhong Tang.
I received my B.S. of Statistics and a dual degree of CS from the School of the Gifted Young, USTC in 2016.
Contact information
Email: pfy824@gmail.com
Wechat: pfy824
Research Interests
My research focuses on machine learning and applications. In particular, I am working on Reinforcement Learning, Uncertainty in AI, and their applications, e.g., quantitative finance, recommendation, and advertising.
Rather than exploiting the knowledge, I prefer exploring intelligence.
Publications
New: our paper “Follow the Prophet: Accurate Online Conversion Rate Prediction in the Face of Delayed Feedback” won the SIGIR 2021 Best Short Paper Honorable Mention award! Congrats to Haoming and our coauthors!
To appear in AAAI 2023.Learn Continuously, Act Discretely: Hybrid Action-Space Reinforcement Learning For Optimal Execution
Feiyang Pan, Tongzhe Zhang, Ling Luo, Jia He and Shuoling Liu.
In IJCAI 2022. [Download paper]Follow the Prophet: Accurate Online Conversion Rate Prediction in the Face of Delayed Feedback
Haoming Li, Feiyang Pan, Xiang Ao, Zhao Yang, Min Lu, Junwei Pan, Dapeng Liu, Lei Xiao, Qing He.
In SIGIR 2021. [Download paper]GuideBoot: Guided Bootstrap for Deep Contextual Bandits
Feiyang Pan, Haoming Li, Xiang Ao, Wei Wang, Yanrong Kang, Ao Tan and Qing He.
In WWW 2021. [Download paper]Trust the Model When It Is Confident: Masked Model-based Actor-Critic
Feiyang Pan, Jia He, Dandan Tu and Qing He.
In NeurIPS 2020. [Download paper]GoChat: Goal Oriented Chatbots with Hierarchical Reinforcement Learning
Jianfeng Liu, Feiyang Pan*, Ling Luo.
In SIGIR 2020. [Download paper]Field-aware calibration: a simple and empirically strong method for reliable probabilistic predictions
Feiyang Pan, Xiang Ao, Pingzhong Tang, Min Lu, Dapeng Liu, Lei Xiao and Qing He.
In WWW 2020. [Download paper] [video]Reading Like HER: Human Reading Inspired Extractive Summarization
Ling Luo, Xiang Ao, Yan Song, Feiyang Pan, Min Yang, Qing He.
In EMNLP 2019. [Download paper]Warm Up Cold-start Advertisements: Improving CTR Predictions via Learning to Learn ID Embeddings
Feiyang Pan, Shuokai Li, Xiang Ao, Pingzhong Tang, Qing He.
In SIGIR 2019. [Download paper] [code]Policy Gradients for Contextual Recommendations
Feiyang Pan, Qingpeng Cai, Pingzhong Tang, Fuzhen Zhuang, Qing He.
In WWW 2019. [Download paper]Policy Optimization with Model-based Explorations
Feiyang Pan, Qingpeng Cai, An-Xiang Zeng, Chun-Xiang Pan, Qing Da, Hualin He, Qing He, Pingzhong Tang.
In AAAI 2019. [Download paper]Beyond Polarity: Interpretable Financial Sentiment Analysis with Hierarchical Query-driven Attention
Ling Luo, Xiang Ao, Feiyang Pan, Jin Wang, Tong Zhao, Ningzi Yu, Qing He.
In IJCAI 2018. [Download paper] [code]
Selected competitions and awards
- (new) NeurIPS-2020: ProcGen competition (on sample efficiency and generalization in reinforcement learning), #1 in round-1, #2 in round-2.
Before diving into research, I was a top-class player in machine learning contests. My highest ranking at Kaggle was 63/90000+.
- Kaggle: Recruit restaurant visitor forecasting, #1 / 2158, 2018.
- Kaggle: TalkingData AdTracking fraud detection challenge, #2 / 3967, 2018.
- IJCAI-18 data mining competition on Alimama sponsored search CVR prediction, #1 / 5204 in season 1, 2018.
- IJCAI-17 data mining competition on customer flow forecasting in Koubei.com, #1 / 4046 in season 1, 2017.
- CCF BDCI 2018 competition on personalized telecom package recommendation, #1 / 2546, 2018.
Some awards in my graduate and undergraduate school:
- XiaPeisu Scholarship, 2021 (the top award in ICT, CAS, only two people get the award every year).
- National Scholarship, 2019.
- LvDe Scholarship (one of the top awards in ICT, CAS), 2019.
- First-class Scholarship of ICT, CAS, 2017.
- Excellent Undergraduate Thesis of USTC, 2016.
- Outstanding Graduate of USTC, 2016.
Academic visits, internships, and industry collaborations
- Ph.D. visitor at Prof. Pingzhong Tang’s group in IIIS, Tsinghua University, 2018-2020.
- Research intern at Machine Learning Group of Ant Financial AI under supervision of Dr. Yuan (Alan) Qi, 2019.
- Research collaboration with Technical Advertising for deep reinforcement learning approaches, 2019-2020.
- Research collaboration with Ant Financial for reinforced anti-fraud mechanisms, 2019-2020.
- Research collaboration with Tencent TEG for advanced methods in advertising, 2018-2019.
- Research collaboration with Alibaba for reinforcement learning in e-commerce, 2018-2019.