Yuandong Tian web metrics

Research Scientist and Senior Manager
Meta AI (FAIR)
Email: yuandong [at] meta [dot] com

Brief Bio

Yuandong Tian is a Research Scientist and Senior Manager in Meta AI Research (FAIR), working on reinforcement learning, representation learning and optimization. He is the first-author recipient of 2021 ICML Outstanding Paper Honorable Mentions and 2013 ICCV Marr Prize Honorable Mentions , and is the lead scientist and engineer for project. He also receives 2022 CGO Distinguished Paper Award . Prior to that, he worked in Google Self-driving Car team in 2013-2014 and received a Ph.D in Robotics Institute, Carnegie Mellon University in 2013.

Google Scholar and CV.

Research Directions: Reinforcement Learning and Optimization, Representation Learning


[Mar. 02, 2023] Invited talk in Microsoft on Long-form Story Generation (, ). Slides here.

[Feb. 28, 2023] Invited talk in IPAM about AI-guided optimization (, ). Slides here.

[Jan. 25, 2023] 1 papers () is accepted in CPAIOR2023!

[Jan. 20, 2023] 3 papers ( ) are accepted in ICLR2023!

[Oct. 17, 2022] 1 papers () is accepted in HPCA2023!

[Oct. 06, 2022] 1 papers () is accepted in EMNLP2022!

[Oct. 03, 2022] Invited talk in MIT Poggio's lab about recent works on contrastive learning (, ). Slides here.

[Sep. 14, 2022] 2 papers ( ) are accepted in NeurIPS2022!

[Sep. 13, 2022] Co-organize AAAI'23 workshop "Reinforcement Learning Ready for Production".

[Aug. 21, 2022] Keynote talk in IEEE Conference on Games.

[Aug. 04, 2022] Invited talk in TTIC workshop on representation learning theory. Link

[Jun. 08, 2022] Invited talk in VALSE Webinar about SSL.

[May. 18, 2022] 1 papers () is accepted in KDD2022!

[May. 15, 2022] 1 papers () is accepted in ICML2022!

[Apr. 28, 2022] Guest lecture in Tianqi Chen's group in CMU.

[Apr. 07, 2022] Invited talk in UIUC about representation learning.

[Mar. 05, 2022] 1 papers () is accepted in CVPR2022!

[Jan. 24, 2022] 3 papers ( ) are accepted in ICLR2022!

[Nov. 29, 2021] 1 papers () is accepted in AAAI2022!

[Nov. 05, 2021] 1 papers () is accepted in CGO2022!

[Sep. 28, 2021] 4 papers ( ) are accepted in NeurIPS2021!

[Jul. 19, 2021] Our paper got ICML Outstanding Paper Award Honorable Mention!

[Jun. 04, 2021] Invited talk in University of Washington NeuralAI Lab about . Slides here. Thanks Eli Shlizerman for inviting!

[May. 08, 2021] 3 papers ( ) are accepted in ICML2021!

[Apr. 29, 2021] 1 papers () is accepted in SIGCOMM2021!

[Apr. 21, 2021] Invited talk in VALSE Webinar about understanding self-supervised learning. Slides here

[Apr. 21, 2021] Invited Guest Lecture in UPenn (Thanks Jing Li) for the invitation. Slides here.

[Apr. 12, 2021] Invited Talk in UCL DARK Lab. Slides here.

[Feb. 28, 2021] 2 papers ( ) are accepted in CVPR2021!

[Jan. 31, 2021] In Black-box optimization challenge of NeurIPS'20, two teams extended our and won 3rd and 8th place! See their reports (JetBrains, KAIST).

[Jan. 22, 2021] 1 papers () is accepted in AIStats2021!

[Dec. 12, 2020] Invited talk in NeurIPS 2020 workshop of Learning meets Combinatorial Algorithms.

[Dec. 12, 2020] Contributed talk in NeurIPS 2020 workshop of Self-supervised Learning, Theory and Practice.

[Nov. 30, 2020] Invited talk (Slides) at Workshop of Reinforcement Learning from Batch Data and Simulation in Simons Institute of UC Berkeley.

[Oct. 20, 2020] Invited Guest lecture in University of Wisconsin Madison (Class syllabus).

[Oct. 14, 2020] Distinguished Guest lecture in IIIS, Tsinghua University.

[Jun. 06, 2020] Invited guest lecture in UCLA.

[Nov. 07, 2019] Invited talk in IAS "Workshop on New Directions in Reinforcement Learning and Control" in Princeton University.

[Nov. 06, 2019] Invited talk in NEC Laboratories Princeton.

[Oct. 27, 2019] Invited talk in AI Sys Workshop in SOSP'19

[Jun. 03, 2019] Long oral talk about in ICML 2019.

[Jan. 03, 2019] Talks in Deep Learning Summit, AAAI 2019 Workshops (Reproducible AI and Game and Environments in Artificial Intelligence).

[Jun. 01, 2018] Multiple talks in Stanford, AI NextCon, etc. link

[Dec. 20, 2017] Keynote at Future Leaders of AI Retreat (FLAIR), Shanghai. Slides here.

[Dec. 06, 2017] Oral talk about platform, NIPS 2017, Long Beach. Slides link.

[Nov. 05, 2017] DRL and Game Tutorial in AI Frontier, Santa Clara. Slides link.

[Oct. 27, 2017] DRL and Game Tutorial in Mountain View, ACMMM 2017. Slides link.

[Aug. 10, 2017] Presentation in Video Games and Machine Learning VGML Workshop, ICML 2017. Slides here. The same talk is also presented in University of Sydney on Aug. 11, hosted by Dong Xu.

[Jul. 03, 2017] On topic "AI In Games: Achievements and Challenges", giving 5 talks in China (CASIA, Tsinghua, Shanghai Tech, Brain-AI workshop and CCF-GAIR 2017) located in Beijing, Shanghai and Shenzhen. Slides here.

Searching Large Neighborhoods for Integer Linear Programs with Contrastive Learning [link]
Taoan Huang, Aaron Ferber, Yuandong Tian, Bistra Dilkina, Benoit Steiner

arXiv 2023

Learning Compiler Pass Orders using Coreset and Normalized Value Prediction [link]
Youwei Liang*, Kevin Stone*, Ali Shameli, Chris Cummins, Mostafa Elhoushi, Jiadong Guo, Benoit Steiner, Xiaomeng Yang, Pengtao Xie, Hugh Leather, Yuandong Tian
(* = Equal 1st authors)

arXiv 2023

Local Branching Relaxation Heuristics for Integer Linear Programs [link]
Taoan Huang, Aaron Ferber, Yuandong Tian, Bistra Dilkina, Benoit Steiner


Efficient Planning in a Compact Latent Action Space [link] [code] [website]
Zhengyao Jiang, Tianjun Zhang, Michael Janner, Yueying Li, Tim Rocktäschel, Edward Grefenstette, Yuandong Tian

ICLR 2023

Understanding the Role of Nonlinearity in Training Dynamics of Contrastive Learning [link] [code] [workshop version] [workshop poster]
Yuandong Tian

ICLR 2023

MACTA: A Multi-agent Reinforcement Learning Approach for Cache Timing Attacks and Detection [link]
Jiaxun Cui, Xiaomeng Yang*, Geunbae Lee*, Mulong Luo*, Peter Stone, Hsien-Hsin S. Lee, Benjamin Lee, G. Edward Suh, Wenjie Xiong**, Yuandong Tian**
(* = Equal 2nd authors, ** = Equal advising)

ICLR 2023

Modeling Scattering Coefficients using Self-Attentive Complex Polynomials with Image-based Representation [link]
Andrew Cohen*, Weiping Dou, Jiang Zhu, Slawomir Koziel, Peter Renner, Jan-Ove Mattsson, Xiaomeng Yang, Beidi Chen, Kevin Stone, Yuandong Tian*
(* = Equal 1st authors)

arXiv 2023

DOC: Improving Long Story Coherence With Detailed Outline Control [link] [code]
Kevin Yang, Dan Klein, Nanyun Peng, Yuandong Tian

arXiv 2022

SurCo: Learning Linear Surrogates For Combinatorial Nonlinear Optimization Problems [link] [Slides]
Aaron Ferber, Taoan Huang, Daochen Zha, Martin Schubert, Benoit Steiner, Bistra Dilkina, Yuandong Tian

arXiv 2022

AutoCAT: Reinforcement Learning for Automated Exploration of Cache Timing-Channel Attacks [link] [code]
Mulong Luo*, Wenjie Xiong*, Geunbae Lee, Yueying Li, Xiaomeng Yang, Amy Zhang, Yuandong Tian, Hsien Hsin S Lee, G Edward Suh
(* = Equal 1st authors)

HPCA 2023

Re3: Generating Longer Stories With Recursive Reprompting and Revision [link] [code]
Kevin Yang, Yuandong Tian, Nanyun Peng, Dan Klein

EMNLP 2022

Understanding Deep Contrastive Learning via Coordinate-wise Optimization [link] [code] [video] [5min talk slides] [poster]
Yuandong Tian

NeurIPS 2022 (Oral)

DreamShard: Generalizable Embedding Table Placement for Recommender Systems [link] [code]
Daochen Zha, Louis Feng, Qiaoyu Tan, Zirui Liu, Kwei-Herng Lai, Bhargav Bhushanam, Yuandong Tian, Arun Kejariwal, Xia Hu

NeurIPS 2022

AutoShard: Automated Embedding Table Sharding for Recommender Systems [link] [code]
Daochen Zha, Louis Feng, Bhargav Bhushanam, Dhruv Choudhary, Jade Nie, Yuandong Tian, Jay Chae, Yinbin Ma, Arun Kejariwal, Xia Hu

KDD 2022

Denoised MDPs: Learning World Models Better Than the World Itself [link] [code] [website]
Tongzhou Wang, Simon S Du, Antonio Torralba, Phillip Isola, Amy Zhang, Yuandong Tian

ICML 2022

On the Importance of Asymmetry for Siamese Representation Learning [link] [code]
Xiao Wang*, Haoqi Fan*, Yuandong Tian, Daisuke Kihara, Xinlei Chen
(* = Equal 1st authors)

CVPR 2022

Understanding Dimensional Collapse in Contrastive Self-supervised Learning [link] [code]
Li Jing, Pascal Vincent, Yann LeCun, Yuandong Tian

ICLR 2022

NASViT: Neural Architecture Search for Efficient Vision Transformers with Gradient Conflict aware Supernet Training [link] [code]
Chengyue Gong, Dilin Wang, Meng Li, Xinlei Chen, Zhicheng Yan, Yuandong Tian, Vikas Chandra

ICLR 2022

Multi-objective Optimization by Learning Space Partitions [link] [code]
Yiyang Zhao, Linnan Wang, Kevin Yang, Tianjun Zhang, Tian Guo, Yuandong Tian

ICLR 2022

Towards demystifying representation learning with non-contrastive self-supervision [link]
Xiang Wang, Xinlei Chen, Simon S Du, Yuandong Tian

arXiv 2021

Sample-Efficient Neural Architecture Search by Learning Action Space [link] [code]
Linnan Wang, Saining Xie, Teng Li, Rodrigo Fonseca, Yuandong Tian

T-PAMI 2021

Learning Bounded Context-Free-Grammar via LSTM and the Transformer: Difference and Explanations [link] [code]
Hui Shi, Sicun Gao, Yuandong Tian, Xinyun Chen, Jishen Zhao

AAAI 2022

CompilerGym: robust, performant compiler optimization environments for AI research [link] [code]
Chris Cummins, Bram Wasti, Jiadong Guo, Brandon Cui, Jason Ansel, Sahir Gomez, Somya Jain, Jia Liu, Olivier Teytaud, Benoit Steiner, Yuandong Tian, Hugh Leather

CGO 2022 (Outstanding Paper)

NovelD: A Simple yet Effective Exploration Criterion [link] [code] [video]
Tianjun Zhang, Huazhe Xu, Xiaolong Wang, Yi Wu, Kurt Keutzer, Joseph E. Gonzalez, Yuandong Tian

NeurIPS 2021

MADE: Exploration via Maximizing Deviation from Explored Regions [link] [code]
Tianjun Zhang*, Paria Rashidinejad*, Jiantao Jiao, Yuandong Tian, Joseph Gonzalez, Stuart Russell
(* = Equal 1st authors)

NeurIPS 2021

Learning Space Partitions for Path Planning [link] [code]
Kevin Yang*, Tianjun Zhang*, Chris Cummins, Brandon Cui, Benoit Steiner, Linnan Wang, Joseph E. Gonzalez, Dan Klein, Yuandong Tian
(* = Equal 1st authors)

NeurIPS 2021

Latent Execution for Neural Program Synthesis Beyond Domain-Specific Languages [link] [code]
Xinyun Chen, Dawn Song, Yuandong Tian

NeurIPS 2021

Understanding Self-supervised Learning Dynamics without Contrastive Pairs [link] [code] [video] [Slides] [Blogpost]
Yuandong Tian, Xinlei Chen, Surya Ganguli

ICML 2021 (Outstanding Paper Award Honorable Mention)

Few-shot Neural Architecture Search [link] [code] [Blogpost]
Yiyang Zhao*, Linnan Wang*, Yuandong Tian, Rodrigo Fonseca, Tian Guo
(* = Equal 1st authors)

ICML 2021 (Long Oral)

Learn-to-Share: A Hardware-friendly Transfer Learning Framework Exploiting Computation and Parameter Sharing [link]
Cheng Fu, Hanxian Huang, Xinyun Chen, Yuandong Tian, Jishen Zhao (UCSD)

ICML 2021 (Long Oral)

Network Planning with Deep Reinforcement Learning [link] [code]
Hang Zhu (JHU), Varun Gupta, Satyajeet Singh Ahuja, Yuandong Tian, Ying Zhang, Xin Jin


FBNetV3: Joint Architecture-Recipe Search using Predictor Pretraining [link]
Xiaoliang Dai, Alvin Wan, Peizhao Zhang, Bichen Wu, Zijian He, Zhen Wei, Kan Chen, Yuandong Tian, Matthew Yu, Peter Vajda, Joseph Gonzalez

CVPR 2021

FPNAS: Fast Probabilistic Neural Architecture Search [link]
Zhicheng Yan, Xiaoliang Dai, Peizhao Zhang, Yuandong Tian, Bichen Wu, Matt Feiszli

CVPR 2021

Understanding Robustness in Teacher-Student Setting: A New Perspective [link] [Slides]
Zhuolin Yang*, Zhaoxi Chen, Tiffany Cai, Xinyun Chen, Bo Li, Yuandong Tian*
(* = Equal 1st authors)

AIStats 2021

Multi-Agent Collaboration via Reward Attribution Decomposition [link] [code] [video] [website]
Tianjun Zhang, Huazhe Xu, Xiaolong Wang, Yi Wu, Kurt Keutzer, Joseph E. Gonzalez, Yuandong Tian

arxiv 2020

Joint Policy Search for Multi-agent Collaboration with Imperfect Information [link] [code] [video]
Yuandong Tian, Qucheng Gong, Tina Jiang

NeurIPS 2020

Understanding Self-supervised Learning with Dual Deep Networks [link] [code] [video]
Yuandong Tian, Lantao Yu, Xinlei Chen, Surya Ganguli

arXiv 2020

Student Specialization in Deep ReLU Networks With Finite Width and Input Dimension [link] [code]
Yuandong Tian

ICML 2020

Playing the lottery with rewards and multiple languages: lottery tickets in RL and NLP [link]
Haonan Yu, Sergey Edunov, Yuandong Tian, Ari S. Morcos

ICLR 2020

Learning Search Space Partition for Black-box Optimization using Monte Carlo Tree Search [link] [code]
Linnan Wang, Rodrigo Fonseca, Yuandong Tian

NeurIPS 2020

Towards Automated Neural Interaction Discovery for Click-Through Rate Prediction [link]
Qingquan Song, Dehua Cheng, Hanning Zhou, Jiyan Yang, Yuandong Tian, Xia Hu

KDD 2020

FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel Dimensions [link] [code]
Alvin Wan, Xiaoliang Dai, Peizhao Zhang, Zijian He, Yuandong Tian, Saining Xie, Bichen Wu, Matthew Yu, Tao Xu, Kan Chen, Peter Vajda, Joseph Gonzalez

CVPR 2020

N-Bref : A High-fidelity Decompiler Exploiting Programming Structures [link] [code] [Blogpost]
Cheng Fu, Kunlin Yang, Xinyun Chen, Yuandong Tian, Jishen Zhao

arxiv 2020

AlphaX: Exploring Neural Architectures with Deep Neural Networks and Monte Carlo Tree Search [link]
Linnan Wang, Yiyang Zhao, Yuu Jinnai, Yuandong Tian, Rodrigo Fonseca

AAAI 2020

Deep Symbolic Superoptimization Without Human Knowledge [link] [code]
Hui Shi, Yang Zhang, Xinyun Chen, Yuandong Tian, Jishen Zhao

ICLR 2020

Hierarchical Decision Making by Generating and Following Natural Language Instructions [link] [code]
Hengyuan Hu*, Denis Yarats*, Qucheng Gong, Yuandong Tian, Mike Lewis
(* = Equal 1st authors)

NeurIPS 2019

ELF OpenGo: An Analysis and Open Reimplementation of AlphaZero [link] [code] [pretrained model and game records] [Other resources] [Blogpost] [Talk]

An open source reimplementation of DeepMind's zero-knowledge training and its application to the game of Go. Trained on 2000 GPUs for 9 days. With a single GPU and 50 seconds per move, the model won 20-0 versus 4 top 30 professional players, given human unlimited thinking time. It also won 980-18 versus LeelaZero (version Apr. 25).

Yuandong Tian, Jerry Ma*, Qucheng Gong*, Shubho Sengupta*, Zhuoyuan Chen, James Pinkerton, Larry Zitnick
(* = Equal 2nd authors)

ICML 2019 (Long Oral)

Algorithmic Framework for Model-based Deep Reinforcement Learning with Theoretical Guarantees [link] [code]
Yuping Luo*, Huazhe Xu*, Yuanzhi Li, Yuandong Tian, Trevor Darrell, Tengyu Ma
(* = Equal 1st authors)

ICLR 2019

M^3RL: Mind-aware Multi-agent Management Reinforcement Learning [link] [code]
Tianmin Shu, Yuandong Tian

ICLR 2019

Luck Matters: Understanding Training Dynamics of Deep ReLU Networks [link] [code] [Poster]
Yuandong Tian, Tina Jiang, Qucheng Gong, Ari Morcos

ICML-workshop 2019

One ticket to win them all: generalizing lottery ticket initializations across datasets and optimizers [link]
Ari S. Morcos, Haonan Yu, Michela Paganini, Yuandong Tian

NeurIPS 2019

Real-world video adaptation with reinforcement learning [link]
Hongzi Mao, Shannon Chen, Drew Dimmery, Shaun Singh, Drew Blaisdell, Yuandong Tian, Mohammad Alizadeh, Eytan Bakshy

ICML-Workshop 2019

FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search [link] [code]
Bichen Wu, Xiaoliang Dai, Peizhao Zhang, Yanghan Wang, Fei Sun, Yiming Wu, Yuandong Tian, Peter Vajda, Yangqing Jia, Kurt Keutzer

CVPR 2019

Learning to Perform Local Rewriting for Combinatorial Optimization [link] [code]
Xinyun Chen, Yuandong Tian

NeurIPS 2019

Coda: An End-to-End Neural Program Decompiler [link]
Cheng Fu, Huili Chen, Haolan Liu, Xinyun Chen, Yuandong Tian, Farinaz Koushanfar, Jishen Zhao

NeurIPS 2019

Building Generalizable Agents with a Realistic and Rich 3D Environment [link] [code]
Yi Wu, Yuxin Wu, Georgia Gkioxari, Yuandong Tian

ICLR-Workshop 2018

A Theoretical Framework for Deep Locally Connected ReLU Network [link] [Poster]
Yuandong Tian

arxiv 2018

Gradient Descent Learns One-hidden-layer CNN: Don't be Afraid of Spurious Local Minima [link]
Simon S. Du, Jason D. Lee, Yuandong Tian, Barnabas Poczos, Aarti Singh

ICML 2018 (Long Oral)

When is a Convolutional Filter Easy To Learn? [link]
Simon S. Du, Jason D. Lee, Yuandong Tian

ICLR 2018

ELF: An Extensive, Lightweight and Flexible Research Platform for Real-time Strategy Games [link] [code] [video]
Yuandong Tian, Qucheng Gong, Wenling Shang, Yuxin Wu, Larry Zitnick

NeurIPS 2017 (Oral)

Training Agent for First-Person Shooter Game with Actor-Critic Curriculum Learning [link]
Yuxin Wu, Yuandong Tian

ICLR 2017

An Analytical Formula of Population Gradient for two-layered ReLU network and its Applications in Convergence and Critical Point Analysis [link] [code]
Yuandong Tian

ICML 2017

Semantic Amodal Segmentation [link]
Yan Zhu, Yuandong Tian, Dimitris Mexatas, Piotr Dollár

CVPR 2017

Better Computer Go Player with Neural Network and Long-term Prediction [link] [code] [pretrained model] [mit tech review]
Yuandong Tian, Yan Zhu

ICLR 2016

Single Image 3D Interpreter Network [link]
Jiajun Wu*, Tianfan Xue*, Joseph J. Lim, Yuandong Tian, Joshua B. Tenenbaum, Antonio Torralba, William T. Freeman
(* = Equal 1st authors)

ECCV 2016 (Oral)

Simple Baseline for Visual Question Answering [link] [code]
Bolei Zhou, Yuandong Tian, Sainbayar Sukhbaatar, Arthur Szlam, Rob Fergus

arxiv 2016

Theory and Practice of Hierarchical Data-driven Descent for Optimal Deformation Estimation [link]
Yuandong Tian, Srinivasa G. Narasimhan

IJCV 2015

Theory and Practice of Globally Optimal Deformation Estimation [link]
Yuandong Tian

PhD thesis 2013

Hierarchical Data-Driven Descent for Efficient Optimal Deformation Estimation [link] [Proofs]
Yuandong Tian, Srinivasa G. Narasimhan

ICCV 2013 (Marr Prize Honorable Mentions)

Integrating Perceptual Learning with External World Knowledge in a Simulated Student [link]
Nan Li, Yuandong Tian, William W. Cohen, Ken Koedinger

AIED 2013

Exploring the Spatial Hierarchy of Mixture Models for Human Pose Estimation [link] [code] [website]
Yuandong Tian, Larry Zitnick, Srinivasa G. Narasimhan

ECCV 2012

Learning from Crowds in the Presence of Schools of Thought [link] [code] [Slides] [Dataset]
Yuandong Tian, Jun Zhu

KDD 2012

Depth from Optical Turbulence [link] [website]
Yuandong Tian, Srinivasa G. Narasimhan, Alan J. Vannevel

CVPR 2012

A Combined Theory of Defocused Illumination and Global Light Transport [link]
Mohit Gupta, Yuandong Tian, Srinivasa G. Narasimhan, Li Zhang

IJCV 2011

Globally Optimal Estimation of Nonrigid Image Distortion [link]
Yuandong Tian, Srinivasa G. Narasimhan

IJCV 2011

Rectification and 3D reconstruction of Curved Document Images [link] [website]
Yuandong Tian, Srinivasa G. Narasimhan

CVPR 2011 (Oral)

Local Isomorphism to Solve the Pre-image Problem in Kernel Methods [link]
Dong Huang, Yuandong Tian, Fernando De la Torre

CVPR 2011

A Globally Optimal Data-Driven Approach for Image Distortion Estimation [link] [website]
Yuandong Tian, Srinivasa G. Narasimhan

CVPR 2010 (Oral)

Seeing through water: Image restoration using model-based tracking [link] [website]
Yuandong Tian, Srinivasa G. Narasimhan

ICCV 2009

(De) Focusing on Global Light Transport for Active Scene Recovery [link]
Mohit Gupta, Yuandong Tian, Srinivasa G. Narasimhan, Li Zhang

CVPR 2009 (Oral)