En‐hui Yang

Modeling and Energy Analysis of Adversarial Perturbations in Deep Image Classification Security

2022

Despite the great success of deep neural networks (DNNs) in computer vision, they are vulnerable to adversarial attacks. Given a well-trained DNN and an image$x$, a malicious and imperceptible perturbation$\varepsilon$can be easily crafted and added …

Conditional Mutual Information Constrained Deep Learning for Classification

2023 · arXiv (Cornell University)

The concepts of conditional mutual information (CMI) and normalized conditional mutual information (NCMI) are introduced to measure the concentration and separation performance of a classification deep neural network (DNN) in the output probability distribution space …

Leveraging Conditional Mutual Information to Improve Large Language Model Fine-Tuning For Classification

2025 · arXiv (Cornell University)

Although large language models (LLMs) have demonstrated remarkable capabilities in recent years, the potential of information theory (IT) to enhance LLM development remains underexplored. This paper introduces the information theoretic principle of Conditional Mutual Information …