Deep neural networks (DNNs) have demonstrated dominating performance in many fields;
since AlexNet, networks used in practice are going wider and deeper. On the theoretical side, …

Gradient inversion attack (or input recovery from gradient) is an emerging threat to the
security and privacy preservation of Federated learning, whereby malicious eavesdroppers or …

Several newly identified tumor suppressor genes including ATM, NBS1, BRCA1 and
BRCA2 are involved in DNA double-strand break repair (DSBR) and DNA damage-induced …

Verifying the robustness property of a general Rectified Linear Unit (ReLU) network is an NP-complete
problem. Although finding the exact minimum adversarial distortion is hard, giving …

Purpose Computer-assisted surgery has been proposed as a technique to improve implant
alignment during total knee arthroplasty (TKA). However, there is still a debate over the …

Mineral scaling is one of the primary factors that constrain the performance of membrane
desalination. However, compared to numerous studies on organic and biological fouling, less …

In this paper, we consider regression problems with one-hidden-layer neural networks (1NNs).
We distill some properties of activation functions that lead to local strong convexity in the …

Large Language Models (LLMs), despite their recent impressive accomplishments, are notably
cost-prohibitive to deploy, particularly for applications involving long-content generation, …

Large language models (LLMs) with hundreds of billions of parameters have sparked a new
wave of exciting AI applications. However, they are computationally expensive at inference …

Ataxia-telangiectasia (AT) and Nijmegen breakage syndrome (NBS) are recessive genetic
disorders with susceptibility to cancer and similar cellular phenotypes 1 . The protein product …