Alert button
Picture for Melanie Ducoffe

Melanie Ducoffe

Alert button

Don't Lie to Me! Robust and Efficient Explainability with Verified Perturbation Analysis

Feb 15, 2022
Thomas Fel, Melanie Ducoffe, David Vigouroux, Remi Cadene, Mikael Capelle, Claire Nicodeme, Thomas Serre

Figure 1 for Don't Lie to Me! Robust and Efficient Explainability with Verified Perturbation Analysis
Figure 2 for Don't Lie to Me! Robust and Efficient Explainability with Verified Perturbation Analysis
Figure 3 for Don't Lie to Me! Robust and Efficient Explainability with Verified Perturbation Analysis
Figure 4 for Don't Lie to Me! Robust and Efficient Explainability with Verified Perturbation Analysis
Viaarxiv icon

Adversarial Active Learning for Deep Networks: a Margin Based Approach

Feb 27, 2018
Melanie Ducoffe, Frederic Precioso

Figure 1 for Adversarial Active Learning for Deep Networks: a Margin Based Approach
Figure 2 for Adversarial Active Learning for Deep Networks: a Margin Based Approach
Figure 3 for Adversarial Active Learning for Deep Networks: a Margin Based Approach
Figure 4 for Adversarial Active Learning for Deep Networks: a Margin Based Approach
Viaarxiv icon

QBDC: Query by dropout committee for training deep supervised architecture

Nov 26, 2015
Melanie Ducoffe, Frederic Precioso

Figure 1 for QBDC: Query by dropout committee for training deep supervised architecture
Figure 2 for QBDC: Query by dropout committee for training deep supervised architecture
Figure 3 for QBDC: Query by dropout committee for training deep supervised architecture
Figure 4 for QBDC: Query by dropout committee for training deep supervised architecture
Viaarxiv icon