(Peter) Xi Chen
About
Right now: I'm building teachable robots at Embodied Intelligence by leveraging recent advances in Deep Imitation Learning and Deep Reinforcement Learning.
----------------------------------------------------------
I’m a PhD student at UC Berkeley in Professor Pieter Abbeel’s group. I was a research scientist at OpenAI working on reinforcement learning and generative models.
My current research interests are generative models and reinforcement learning, which try to endow machines with the abilities to understand and act in complicated environments respectively.
I co-organized the NIPS 2016 Deep RL Workshop.
I will give a tutorial on Deep RL at 2017 PKU Summer School on Data Science.
Research (see google scholar for an up-to-date list)
Pre-prints
- PixelSNAIL: An Improved Autoregressive Generative Model
Xi Chen, Nikhil Mishra, Mostafa Rohaninejad, Pieter Abbeel
[ArXiv] - Safer Classification by Synthesis
William Wang, Angelina Wang, Aviv Tamar, Xi Chen, Pieter Abbeel
[ArXiv] - Equivalence Between Policy Gradients and Soft Q-Learning
John Schulman, Xi Chen, Pieter Abbeel
[ArXiv] - Evolution strategies as a scalable alternative to reinforcement learning
Tim Salimans, Jonathan Ho, Xi Chen, Ilya Sutskever
[ArXiv, Code, Blog post] - RL2: Fast Reinforcement Learning via Slow Reinforcement Learning
Yan Duan, John Schulman, Xi Chen, Peter L. Bartlett, Ilya Sutskever, Pieter Abbeel
[ArXiv]
Publications
- A Simple Neural Attentive Meta-Learner
Nikhil Mishra*, Mostafa Rohaninejad*, Xi Chen, Pieter Abbeel
ICLR 2018 [ArXiv] - Meta Learning Shared Hierarchies
Kevin Frans, Jonathan Ho, Xi Chen, Pieter Abbeel, John Schulman
ICLR 2018 [ArXiv] - Parameter Space Noise for Exploration
Matthias Plappert, Rein Houthooft, Prafulla Dhariwal, Szymon Sidor, Richard Y. Chen, Xi Chen, Tamim Asfour, Pieter Abbeel, Marcin Andrychowicz
ICLR 2018 [ArXiv] - #Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning
Haoran Tang, Rein Houthooft, Davis Foote, Adam Stooke, Xi Chen, Yan Duan, John Schulman, Filip De Turck, Pieter Abbeel
NIPS 2017 [ArXiv] - Variational Lossy Autoencoder
Xi Chen, Diederik P. Kingma, Tim Salimans, Yan Duan, Prafulla Dhariwal, John Schulman, Ilya Sutskever, Pieter Abbeel
ICLR 2017 [ArXiv] - PixelCNN++: Improving the PixelCNN with Discretized Logistic Mixture Likelihood and Other Modifications
Tim Salimans, Andrej Karpathy, Xi Chen, Diederik P. Kingma
ICLR 2017 [ArXiv, Code] - InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
Xi Chen, Yan Duan, Rein Houthooft, John Schulman, Ilya Sutskever, Pieter Abbeel
NIPS 2016 [ArXiv, Code, Blog post]
Invited oral presentation at the NIPS 2016 Deep Learning Symposium - VIME: Variational Information Maximizing Exploration
Rein Houthooft, Xi Chen, Yan Duan, John Schulman, Filip De Turck, Pieter Abbeel
NIPS 2016 [ArXiv, Code, Blog post] - Improving Variational Autoencoders with Inverse Autoregressive Flow
Diederik P Kingma, Tim Salimans, Rafal Jozefowicz, Xi Chen, Ilya Sutskever, Max Welling
NIPS 2016 [Paper, Code, Blog post]
Invited poster presentation at the NIPS 2016 Deep Learning Symposium - Improved Techniques for Training GANs
Tim Salimans, Ian Goodfellow, Wojciech Zaremba, Vicki Cheung, Alec Radford, Xi Chen
NIPS 2016 [ArXiv, Code, Blog post] - Benchmarking Deep Reinforcement Learning for Continuous Control
Yan Duan, Xi Chen, Rein Houthooft, John Schulman, Pieter Abbeel
ICML 2016 [ArXiv, Code, Slides]
Older Work
DIABLO: A Warehouse-Scale Computer Network Simulator using FPGAs
Zhangxi Tan, Zhenghao Qian, Xi Chen, Krste Asanovic, David Patterson
ASPLOS 2015 [Website]Evaluating computational models of explanation using human judgments
Michael Pacer, Joseph Williams, Xi Chen, Tania Lombrozo, Thomas Griffiths
UAI 2013 [ArXiv, Code]
Contact
Feel free to contact me at