Add to ORKGWe propose to utilize gradients for detecting adversarial and out-of-distribution samples. We introduce confounding labels -- labels that differ from normal labels seen during training -- in gradient generation to probe the effective expressivity of neural networks. Gradients depict the amount of change required for a model to properly represent given inputs, providing insight into the representational power of the model established by network architectural properties as well as training data. By introducing a label of different design, we remove the dependency on ground truth labels for gradient generation during inference. We show that our gradient-based approach allows for capturing the anomaly in inputs based on the effective expressivi...
In this work, we provide a characterization of the feature-learning process in two-layer ReLU networ...
Deep learning plays an important role in various disciplines, such as auto-driving, information tech...
Despite impressive success in many tasks, deep learning models are shown to rely on spurious feature...
One critical challenge in deploying highly performant machine learning models in real-life applicati...
In adversarial examples, humans can easily classify the images even though the images are corrupted...
Adversarial attacks and defenses are currently active areas of research for the deep learning commun...
Current machine learning models achieve super-human performance in many real-world applications. Sti...
We consider distributed (gradient descent-based) learning scenarios where the server combines the gr...
Deep neural networks have proven remarkably effective at solving many classification problems, but h...
Recent advancements in the field of deep learning have substantially increased the adoption rate of ...
Despite the widespread use of machine learning in adversarial settings such as computer security, re...
One of the primary challenges limiting the applicability of deep learning is its susceptibility to l...
Recent studies show that the deep neural networks (DNNs) have achieved great success in various task...
Modern neural networks can assign high confidence to inputs drawn from outside the training distribu...
Deep neural networks are known to be vulnerable to adversarial examples crafted by adding human-impe...
In this work, we provide a characterization of the feature-learning process in two-layer ReLU networ...
Deep learning plays an important role in various disciplines, such as auto-driving, information tech...
Despite impressive success in many tasks, deep learning models are shown to rely on spurious feature...
One critical challenge in deploying highly performant machine learning models in real-life applicati...
In adversarial examples, humans can easily classify the images even though the images are corrupted...
Adversarial attacks and defenses are currently active areas of research for the deep learning commun...
Current machine learning models achieve super-human performance in many real-world applications. Sti...
We consider distributed (gradient descent-based) learning scenarios where the server combines the gr...
Deep neural networks have proven remarkably effective at solving many classification problems, but h...
Recent advancements in the field of deep learning have substantially increased the adoption rate of ...
Despite the widespread use of machine learning in adversarial settings such as computer security, re...
One of the primary challenges limiting the applicability of deep learning is its susceptibility to l...
Recent studies show that the deep neural networks (DNNs) have achieved great success in various task...
Modern neural networks can assign high confidence to inputs drawn from outside the training distribu...
Deep neural networks are known to be vulnerable to adversarial examples crafted by adding human-impe...
In this work, we provide a characterization of the feature-learning process in two-layer ReLU networ...
Deep learning plays an important role in various disciplines, such as auto-driving, information tech...
Despite impressive success in many tasks, deep learning models are shown to rely on spurious feature...