Picture for Nimit S. Sohoni

Nimit S. Sohoni

Correct-N-Contrast: A Contrastive Approach for Improving Robustness to Spurious Correlations

Add code
Mar 03, 2022
Figure 1 for Correct-N-Contrast: A Contrastive Approach for Improving Robustness to Spurious Correlations
Figure 2 for Correct-N-Contrast: A Contrastive Approach for Improving Robustness to Spurious Correlations
Figure 3 for Correct-N-Contrast: A Contrastive Approach for Improving Robustness to Spurious Correlations
Figure 4 for Correct-N-Contrast: A Contrastive Approach for Improving Robustness to Spurious Correlations
Viaarxiv icon

Low-Shot Validation: Active Importance Sampling for Estimating Classifier Performance on Rare Categories

Add code
Sep 13, 2021
Figure 1 for Low-Shot Validation: Active Importance Sampling for Estimating Classifier Performance on Rare Categories
Figure 2 for Low-Shot Validation: Active Importance Sampling for Estimating Classifier Performance on Rare Categories
Figure 3 for Low-Shot Validation: Active Importance Sampling for Estimating Classifier Performance on Rare Categories
Figure 4 for Low-Shot Validation: Active Importance Sampling for Estimating Classifier Performance on Rare Categories
Viaarxiv icon

Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps

Add code
Jan 05, 2021
Figure 1 for Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps
Figure 2 for Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps
Figure 3 for Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps
Figure 4 for Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps
Viaarxiv icon

No Subclass Left Behind: Fine-Grained Robustness in Coarse-Grained Classification Problems

Add code
Nov 25, 2020
Figure 1 for No Subclass Left Behind: Fine-Grained Robustness in Coarse-Grained Classification Problems
Figure 2 for No Subclass Left Behind: Fine-Grained Robustness in Coarse-Grained Classification Problems
Figure 3 for No Subclass Left Behind: Fine-Grained Robustness in Coarse-Grained Classification Problems
Figure 4 for No Subclass Left Behind: Fine-Grained Robustness in Coarse-Grained Classification Problems
Viaarxiv icon