Picture for Robbie Jones

Robbie Jones

Connect, Not Collapse: Explaining Contrastive Learning for Unsupervised Domain Adaptation

Add code
Apr 01, 2022
Figure 1 for Connect, Not Collapse: Explaining Contrastive Learning for Unsupervised Domain Adaptation
Figure 2 for Connect, Not Collapse: Explaining Contrastive Learning for Unsupervised Domain Adaptation
Figure 3 for Connect, Not Collapse: Explaining Contrastive Learning for Unsupervised Domain Adaptation
Figure 4 for Connect, Not Collapse: Explaining Contrastive Learning for Unsupervised Domain Adaptation
Viaarxiv icon

Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution

Add code
Feb 21, 2022
Figure 1 for Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution
Figure 2 for Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution
Figure 3 for Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution
Figure 4 for Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distribution
Viaarxiv icon

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

Add code
Dec 16, 2020
Figure 1 for In-N-Out: Pre-Training and Self-Training using Auxiliary Information for Out-of-Distribution Robustness
Figure 2 for In-N-Out: Pre-Training and Self-Training using Auxiliary Information for Out-of-Distribution Robustness
Figure 3 for In-N-Out: Pre-Training and Self-Training using Auxiliary Information for Out-of-Distribution Robustness
Figure 4 for In-N-Out: Pre-Training and Self-Training using Auxiliary Information for Out-of-Distribution Robustness
Viaarxiv icon