4 papers
Improving LIME Robustness with Smarter Locality Sampling
Sean Saito, Eugene Chua, Nicholas Capel +1
Explainability algorithms such as LIME have enabled machine learning systems to adopt transparency and fairness, which are important qualities in commercial use cases. However, rec…
Explaining Away Attacks Against Neural Networks
Sean Saito, Jin Wang
We investigate the problem of identifying adversarial attacks on image-based neural networks. We present intriguing experimental results showing significant discrepancies between t…
Effects of Loss Functions And Target Representations on Adversarial Robustness
Sean Saito, Sujoy Roy
Understanding and evaluating the robustness of neural networks under adversarial settings is a subject of growing interest. Attacks proposed in the literature usually work with mod…
HiNet: Hierarchical Classification with Neural Network
Zhenzhou Wu, Sean Saito
Traditionally, classifying large hierarchical labels with more than 10000 distinct traces can only be achieved with flatten labels. Although flatten labels is feasible, it misses t…