Picture for Joshua Lockhart

Joshua Lockhart

Learn to explain yourself, when you can: Equipping Concept Bottleneck Models with the ability to abstain on their concept predictions

Add code
Nov 21, 2022
Viaarxiv icon

Towards learning to explain with concept bottleneck models: mitigating information leakage

Add code
Nov 07, 2022
Viaarxiv icon

Feature Importance for Time Series Data: Improving KernelSHAP

Add code
Oct 05, 2022
Figure 1 for Feature Importance for Time Series Data: Improving KernelSHAP
Figure 2 for Feature Importance for Time Series Data: Improving KernelSHAP
Viaarxiv icon

Reductive MDPs: A Perspective Beyond Temporal Horizons

Add code
May 15, 2022
Figure 1 for Reductive MDPs: A Perspective Beyond Temporal Horizons
Figure 2 for Reductive MDPs: A Perspective Beyond Temporal Horizons
Figure 3 for Reductive MDPs: A Perspective Beyond Temporal Horizons
Figure 4 for Reductive MDPs: A Perspective Beyond Temporal Horizons
Viaarxiv icon

Asynchronous Collaborative Learning Across Data Silos

Add code
Mar 23, 2022
Figure 1 for Asynchronous Collaborative Learning Across Data Silos
Figure 2 for Asynchronous Collaborative Learning Across Data Silos
Figure 3 for Asynchronous Collaborative Learning Across Data Silos
Figure 4 for Asynchronous Collaborative Learning Across Data Silos
Viaarxiv icon

SURF: Improving classifiers in production by learning from busy and noisy end users

Add code
Oct 12, 2020
Figure 1 for SURF: Improving classifiers in production by learning from busy and noisy end users
Figure 2 for SURF: Improving classifiers in production by learning from busy and noisy end users
Viaarxiv icon

Some people aren't worth listening to: periodically retraining classifiers with feedback from a team of end users

Add code
Apr 27, 2020
Figure 1 for Some people aren't worth listening to: periodically retraining classifiers with feedback from a team of end users
Figure 2 for Some people aren't worth listening to: periodically retraining classifiers with feedback from a team of end users
Figure 3 for Some people aren't worth listening to: periodically retraining classifiers with feedback from a team of end users
Figure 4 for Some people aren't worth listening to: periodically retraining classifiers with feedback from a team of end users
Viaarxiv icon