Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Tengyu Ma

Learning Barrier Certificates: Towards Safe Reinforcement Learning with Zero Training-time Violations


Aug 04, 2021
Yuping Luo, Tengyu Ma


  Access Paper or Ask Questions

Statistically Meaningful Approximation: a Case Study on Approximating Turing Machines with Transformers


Jul 28, 2021
Colin Wei, Yining Chen, Tengyu Ma


  Access Paper or Ask Questions

Calibrating Predictions to Decisions: A Novel Approach to Multi-Class Calibration


Jul 12, 2021
Shengjia Zhao, Michael P. Kim, Roshni Sahoo, Tengyu Ma, Stefano Ermon


  Access Paper or Ask Questions

Iterative Feature Matching: Toward Provable Domain Generalization with Logarithmic Environments


Jun 18, 2021
Yining Chen, Elan Rosenfeld, Mark Sellke, Tengyu Ma, Andrej Risteski


  Access Paper or Ask Questions

Why Do Pretrained Language Models Help in Downstream Tasks? An Analysis of Head and Prompt Tuning


Jun 17, 2021
Colin Wei, Sang Michael Xie, Tengyu Ma


  Access Paper or Ask Questions

Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Loss


Jun 17, 2021
Jeff Z. HaoChen, Colin Wei, Adrien Gaidon, Tengyu Ma


  Access Paper or Ask Questions

Label Noise SGD Provably Prefers Flat Global Minimizers


Jun 11, 2021
Alex Damian, Tengyu Ma, Jason Lee

* 53 pages, 4 figures, under review for NeurIPS 2021 

  Access Paper or Ask Questions

Joint System-Wise Optimization for Pipeline Goal-Oriented Dialog System


Jun 09, 2021
Zichuan Lin, Jing Huang, Bowen Zhou, Xiaodong He, Tengyu Ma

* 13 pages 

  Access Paper or Ask Questions

Why Do Local Methods Solve Nonconvex Problems?


Mar 24, 2021
Tengyu Ma

* This is the Chapter 21 of the book "Beyond the Worst-Case Analysis of Algorithms" 

  Access Paper or Ask Questions

Fine-Grained Gap-Dependent Bounds for Tabular MDPs via Adaptive Multi-Step Bootstrap


Feb 09, 2021
Haike Xu, Tengyu Ma, Simon S. Du


  Access Paper or Ask Questions

Provable Model-based Nonlinear Bandit and Reinforcement Learning: Shelve Optimism, Embrace Virtual Curvature


Feb 08, 2021
Kefan Dong, Jiaqi Yang, Tengyu Ma


  Access Paper or Ask Questions

In-N-Out: Pre-Training and Self-Training using Auxiliary Information for Out-of-Distribution Robustness


Dec 16, 2020
Sang Michael Xie, Ananya Kumar, Robbie Jones, Fereshte Khani, Tengyu Ma, Percy Liang


  Access Paper or Ask Questions

Meta-learning Transferable Representations with a Single Target Domain


Nov 03, 2020
Hong Liu, Jeff Z. HaoChen, Colin Wei, Tengyu Ma


  Access Paper or Ask Questions

Document-Level Relation Extraction with Adaptive Thresholding and Localized Context Pooling


Oct 23, 2020
Wenxuan Zhou, Kevin Huang, Tengyu Ma, Jing Huang

* Code available at https://github.com/wzhouad/ATLOP 

  Access Paper or Ask Questions

Beyond Lazy Training for Over-parameterized Tensor Decomposition


Oct 22, 2020
Xiang Wang, Chenwei Wu, Jason D. Lee, Tengyu Ma, Rong Ge

* NeurIPS 2020; the first two authors contribute equally 

  Access Paper or Ask Questions

Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data


Oct 15, 2020
Colin Wei, Kendrick Shen, Yining Chen, Tengyu Ma


  Access Paper or Ask Questions

Entity and Evidence Guided Relation Extraction for DocRED


Aug 27, 2020
Kevin Huang, Guangtao Wang, Tengyu Ma, Jing Huang


  Access Paper or Ask Questions

Learning Over-Parametrized Two-Layer ReLU Neural Networks beyond NTK


Jul 09, 2020
Yuanzhi Li, Tengyu Ma, Hongyang R. Zhang

* Conference on Learning Theory (COLT) 2020 

  Access Paper or Ask Questions

Individual Calibration with Randomized Forecasting


Jul 05, 2020
Shengjia Zhao, Tengyu Ma, Stefano Ermon


  Access Paper or Ask Questions

Simplifying Models with Unlabeled Output Data


Jun 29, 2020
Sang Michael Xie, Tengyu Ma, Percy Liang


  Access Paper or Ask Questions

Heteroskedastic and Imbalanced Deep Learning with Adaptive Regularization


Jun 29, 2020
Kaidi Cao, Yining Chen, Junwei Lu, Nikos Arechiga, Adrien Gaidon, Tengyu Ma


  Access Paper or Ask Questions

Active Online Domain Adaptation


Jun 25, 2020
Yining Chen, Haipeng Luo, Tengyu Ma, Chicheng Zhang


  Access Paper or Ask Questions

Federated Accelerated Stochastic Gradient Descent


Jun 19, 2020
Honglin Yuan, Tengyu Ma

* 76 pages, 7 figures, 2 tables 

  Access Paper or Ask Questions

Shape Matters: Understanding the Implicit Bias of the Noise Covariance


Jun 18, 2020
Jeff Z. HaoChen, Colin Wei, Jason D. Lee, Tengyu Ma


  Access Paper or Ask Questions

Self-training Avoids Using Spurious Features Under Domain Shift


Jun 17, 2020
Yining Chen, Colin Wei, Ananya Kumar, Tengyu Ma


  Access Paper or Ask Questions