Alert button
Picture for Thang Doan

Thang Doan

Alert button

GradOrth: A Simple yet Efficient Out-of-Distribution Detection with Orthogonal Projection of Gradients

Aug 01, 2023
Sima Behpour, Thang Doan, Xin Li, Wenbin He, Liang Gou, Liu Ren

Figure 1 for GradOrth: A Simple yet Efficient Out-of-Distribution Detection with Orthogonal Projection of Gradients
Figure 2 for GradOrth: A Simple yet Efficient Out-of-Distribution Detection with Orthogonal Projection of Gradients
Figure 3 for GradOrth: A Simple yet Efficient Out-of-Distribution Detection with Orthogonal Projection of Gradients
Figure 4 for GradOrth: A Simple yet Efficient Out-of-Distribution Detection with Orthogonal Projection of Gradients
Viaarxiv icon

UP-DP: Unsupervised Prompt Learning for Data Pre-Selection with Vision-Language Models

Jul 20, 2023
Xin Li, Sima Behpour, Thang Doan, Wenbin He, Liang Gou, Liu Ren

Figure 1 for UP-DP: Unsupervised Prompt Learning for Data Pre-Selection with Vision-Language Models
Figure 2 for UP-DP: Unsupervised Prompt Learning for Data Pre-Selection with Vision-Language Models
Figure 3 for UP-DP: Unsupervised Prompt Learning for Data Pre-Selection with Vision-Language Models
Figure 4 for UP-DP: Unsupervised Prompt Learning for Data Pre-Selection with Vision-Language Models
Viaarxiv icon

Hyp-OW: Exploiting Hierarchical Structure Learning with Hyperbolic Distance Enhances Open World Object Detection

Jun 25, 2023
Thang Doan, Xin Li, Sima Behpour, Wenbin He, Liang Gou, Liu Ren

Figure 1 for Hyp-OW: Exploiting Hierarchical Structure Learning with Hyperbolic Distance Enhances Open World Object Detection
Figure 2 for Hyp-OW: Exploiting Hierarchical Structure Learning with Hyperbolic Distance Enhances Open World Object Detection
Figure 3 for Hyp-OW: Exploiting Hierarchical Structure Learning with Hyperbolic Distance Enhances Open World Object Detection
Figure 4 for Hyp-OW: Exploiting Hierarchical Structure Learning with Hyperbolic Distance Enhances Open World Object Detection
Viaarxiv icon

Building a Subspace of Policies for Scalable Continual Learning

Nov 18, 2022
Jean-Baptiste Gaya, Thang Doan, Lucas Caccia, Laure Soulier, Ludovic Denoyer, Roberta Raileanu

Figure 1 for Building a Subspace of Policies for Scalable Continual Learning
Figure 2 for Building a Subspace of Policies for Scalable Continual Learning
Figure 3 for Building a Subspace of Policies for Scalable Continual Learning
Figure 4 for Building a Subspace of Policies for Scalable Continual Learning
Viaarxiv icon

Efficient Continual Learning Ensembles in Neural Network Subspaces

Feb 20, 2022
Thang Doan, Seyed Iman Mirzadeh, Joelle Pineau, Mehrdad Farajtabar

Figure 1 for Efficient Continual Learning Ensembles in Neural Network Subspaces
Figure 2 for Efficient Continual Learning Ensembles in Neural Network Subspaces
Figure 3 for Efficient Continual Learning Ensembles in Neural Network Subspaces
Figure 4 for Efficient Continual Learning Ensembles in Neural Network Subspaces
Viaarxiv icon

Domain Adversarial Reinforcement Learning

Feb 14, 2021
Bonnie Li, Vincent François-Lavet, Thang Doan, Joelle Pineau

Figure 1 for Domain Adversarial Reinforcement Learning
Figure 2 for Domain Adversarial Reinforcement Learning
Figure 3 for Domain Adversarial Reinforcement Learning
Figure 4 for Domain Adversarial Reinforcement Learning
Viaarxiv icon

Regularized Inverse Reinforcement Learning

Oct 07, 2020
Wonseok Jeon, Chen-Yang Su, Paul Barde, Thang Doan, Derek Nowrouzezahrai, Joelle Pineau

Figure 1 for Regularized Inverse Reinforcement Learning
Figure 2 for Regularized Inverse Reinforcement Learning
Figure 3 for Regularized Inverse Reinforcement Learning
Figure 4 for Regularized Inverse Reinforcement Learning
Viaarxiv icon

A Theoretical Analysis of Catastrophic Forgetting through the NTK Overlap Matrix

Oct 07, 2020
Thang Doan, Mehdi Bennani, Bogdan Mazoure, Guillaume Rabusseau, Pierre Alquier

Figure 1 for A Theoretical Analysis of Catastrophic Forgetting through the NTK Overlap Matrix
Figure 2 for A Theoretical Analysis of Catastrophic Forgetting through the NTK Overlap Matrix
Figure 3 for A Theoretical Analysis of Catastrophic Forgetting through the NTK Overlap Matrix
Figure 4 for A Theoretical Analysis of Catastrophic Forgetting through the NTK Overlap Matrix
Viaarxiv icon

Deep Reinforcement and InfoMax Learning

Jun 12, 2020
Bogdan Mazoure, Remi Tachet des Combes, Thang Doan, Philip Bachman, R Devon Hjelm

Figure 1 for Deep Reinforcement and InfoMax Learning
Figure 2 for Deep Reinforcement and InfoMax Learning
Figure 3 for Deep Reinforcement and InfoMax Learning
Figure 4 for Deep Reinforcement and InfoMax Learning
Viaarxiv icon

Provably efficient reconstruction of policy networks

Feb 07, 2020
Bogdan Mazoure, Thang Doan, Tianyu Li, Vladimir Makarenkov, Joelle Pineau, Doina Precup, Guillaume Rabusseau

Figure 1 for Provably efficient reconstruction of policy networks
Figure 2 for Provably efficient reconstruction of policy networks
Figure 3 for Provably efficient reconstruction of policy networks
Figure 4 for Provably efficient reconstruction of policy networks
Viaarxiv icon