Alert button
Picture for Tengyu Ma

Tengyu Ma

Alert button

Learning Self-Correctable Policies and Value Functions from Demonstrations with Negative Sampling

Aug 01, 2019
Yuping Luo, Huazhe Xu, Tengyu Ma

Figure 1 for Learning Self-Correctable Policies and Value Functions from Demonstrations with Negative Sampling
Figure 2 for Learning Self-Correctable Policies and Value Functions from Demonstrations with Negative Sampling
Figure 3 for Learning Self-Correctable Policies and Value Functions from Demonstrations with Negative Sampling
Viaarxiv icon

A Model-based Approach for Sample-efficient Multi-task Reinforcement Learning

Jul 15, 2019
Nicholas C. Landolfi, Garrett Thomas, Tengyu Ma

Figure 1 for A Model-based Approach for Sample-efficient Multi-task Reinforcement Learning
Figure 2 for A Model-based Approach for Sample-efficient Multi-task Reinforcement Learning
Figure 3 for A Model-based Approach for Sample-efficient Multi-task Reinforcement Learning
Viaarxiv icon

Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks

Jul 10, 2019
Yuanzhi Li, Colin Wei, Tengyu Ma

Figure 1 for Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks
Figure 2 for Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks
Figure 3 for Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks
Figure 4 for Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks
Viaarxiv icon

Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss

Jun 18, 2019
Kaidi Cao, Colin Wei, Adrien Gaidon, Nikos Arechiga, Tengyu Ma

Figure 1 for Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss
Figure 2 for Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss
Figure 3 for Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss
Figure 4 for Learning Imbalanced Datasets with Label-Distribution-Aware Margin Loss
Viaarxiv icon

Data-dependent Sample Complexity of Deep Neural Networks via Lipschitz Augmentation

May 29, 2019
Colin Wei, Tengyu Ma

Figure 1 for Data-dependent Sample Complexity of Deep Neural Networks via Lipschitz Augmentation
Figure 2 for Data-dependent Sample Complexity of Deep Neural Networks via Lipschitz Augmentation
Figure 3 for Data-dependent Sample Complexity of Deep Neural Networks via Lipschitz Augmentation
Viaarxiv icon

On the Performance of Thompson Sampling on Logistic Bandits

May 12, 2019
Shi Dong, Tengyu Ma, Benjamin Van Roy

Figure 1 for On the Performance of Thompson Sampling on Logistic Bandits
Viaarxiv icon

Fixup Initialization: Residual Learning Without Normalization

Jan 27, 2019
Hongyi Zhang, Yann N. Dauphin, Tengyu Ma

Figure 1 for Fixup Initialization: Residual Learning Without Normalization
Figure 2 for Fixup Initialization: Residual Learning Without Normalization
Figure 3 for Fixup Initialization: Residual Learning Without Normalization
Figure 4 for Fixup Initialization: Residual Learning Without Normalization
Viaarxiv icon