Picture for Laura Rieger

Laura Rieger

A simple defense against adversarial attacks on heatmap explanations

Add code
Jul 13, 2020
Figure 1 for A simple defense against adversarial attacks on heatmap explanations
Figure 2 for A simple defense against adversarial attacks on heatmap explanations
Figure 3 for A simple defense against adversarial attacks on heatmap explanations
Figure 4 for A simple defense against adversarial attacks on heatmap explanations
Viaarxiv icon

Client Adaptation improves Federated Learning with Simulated Non-IID Clients

Add code
Jul 09, 2020
Figure 1 for Client Adaptation improves Federated Learning with Simulated Non-IID Clients
Figure 2 for Client Adaptation improves Federated Learning with Simulated Non-IID Clients
Viaarxiv icon

IROF: a low resource evaluation metric for explanation methods

Add code
Mar 09, 2020
Figure 1 for IROF: a low resource evaluation metric for explanation methods
Figure 2 for IROF: a low resource evaluation metric for explanation methods
Figure 3 for IROF: a low resource evaluation metric for explanation methods
Figure 4 for IROF: a low resource evaluation metric for explanation methods
Viaarxiv icon

Interpretations are useful: penalizing explanations to align neural networks with prior knowledge

Add code
Oct 01, 2019
Figure 1 for Interpretations are useful: penalizing explanations to align neural networks with prior knowledge
Figure 2 for Interpretations are useful: penalizing explanations to align neural networks with prior knowledge
Figure 3 for Interpretations are useful: penalizing explanations to align neural networks with prior knowledge
Figure 4 for Interpretations are useful: penalizing explanations to align neural networks with prior knowledge
Viaarxiv icon

Aggregating explainability methods for neural networks stabilizes explanations

Add code
Mar 01, 2019
Figure 1 for Aggregating explainability methods for neural networks stabilizes explanations
Figure 2 for Aggregating explainability methods for neural networks stabilizes explanations
Figure 3 for Aggregating explainability methods for neural networks stabilizes explanations
Figure 4 for Aggregating explainability methods for neural networks stabilizes explanations
Viaarxiv icon