Found 15510 results, showing the newest relevant preprints. Sort by relevancy only.Update me on new preprints

Networks of HIV-1 envelope glycans maintain antibody epitopes in the face of glycan additions and deletions

Glycan networks underpin the conservation of bnAb epitopes and are an important parameter in immunogen design. Expand abstract.
1 day ago
7/10 relevant
bioRxiv

Improving the Tightness of Convex Relaxation Bounds for Training Certifiably Robust Classifiers

We propose two regularizers that can be used to train neural networks that yield tighter convex relaxation bounds for robustness. Expand abstract.
2 days ago
4/10 relevant
arXiv

Non-Intrusive Detection of Adversarial Deep Learning Attacks via Observer Networks

The use of multiple observer networks makes attacking the detection mechanism non-trivial even when the attacker is aware of the victim classifier. Expand abstract.
2 days ago
9/10 relevant
arXiv

Convex Duality of Deep Neural Networks

As a corollary, we prove that norm regularized deep ReLU networks yield spline interpolation for one-dimensional datasets which was previously known only for two-layer networks. Expand abstract.
2 days ago
9/10 relevant
arXiv

Detection and Classification of Astronomical Targets with Deep Neural Networks in Wide Field Small Aperture Telescopes

In this paper, we propose an astronomical targets detection and classification framework based on deep neural networks for images obtained by wide field small aperture telescopes. Expand abstract.
3 days ago
7/10 relevant
arXiv

Exploiting the Full Capacity of Deep Neural Networks while Avoiding Overfitting by Targeted Sparsity Regularization

Overfitting is one of the most common problems when training deep neural networks on comparatively small datasets. Expand abstract.
3 days ago
7/10 relevant
arXiv

Learning to Simulate Complex Physics with Graph Networks

Our framework---which we term "Graph Network-based Simulators" (GNS)---represents the state of a physical system with particles, expressed as nodes in a graph, and computes dynamics via learned message-passing. Expand abstract.
3 days ago
5/10 relevant
arXiv

Conditional Independence in Max-linear Bayesian Networks

Motivated by extreme value theory, max-linear Bayesian networks have been recently introduced and studied as an alternative to linear structural equation models. Expand abstract.
3 days ago
9/10 relevant
arXiv

Towards Robust and Reproducible Active Learning Using Neural Networks

We believe our findings and recommendations will help advance reproducible research in robust AL using neural networks. Expand abstract.
3 days ago
8/10 relevant
arXiv

Photorealistic Lip Sync with Adversarial Temporal Convolutional Networks

We model the seq-to-seq mapping from audio signals to mouth features by two adversarial temporal convolutional networks. Expand abstract.
4 days ago
6/10 relevant
arXiv