Machine learning in cybersecurity: a comprehensive survey

D Dasgupta, Z Akhtar, S Sen - The Journal of Defense …, 2022‏ - journals.sagepub.com
Today's world is highly network interconnected owing to the pervasiveness of small personal
devices (eg, smartphones) as well as large computing devices or services (eg, cloud …

Comprehensive survey of deep learning in remote sensing: theories, tools, and challenges for the community

JE Ball, DT Anderson, CS Chan - Journal of applied remote …, 2017‏ - spiedigitallibrary.org
In recent years, deep learning (DL), a rebranding of neural networks (NNs), has risen to the
top in numerous areas, namely computer vision (CV), speech recognition, and natural …

Simple and principled uncertainty estimation with deterministic deep learning via distance awareness

J Liu, Z Lin, S Padhy, D Tran… - Advances in neural …, 2020‏ - proceedings.neurips.cc
Bayesian neural networks (BNN) and deep ensembles are principled approaches to
estimate the predictive uncertainty of a deep learning model. However their practicality in …

Warm: On the benefits of weight averaged reward models

A Ramé, N Vieillard, L Hussenot, R Dadashi… - arxiv preprint arxiv …, 2024‏ - arxiv.org
Aligning large language models (LLMs) with human preferences through reinforcement
learning (RLHF) can lead to reward hacking, where LLMs exploit failures in the reward …

Wild patterns: Ten years after the rise of adversarial machine learning

B Biggio, F Roli - Proceedings of the 2018 ACM SIGSAC Conference on …, 2018‏ - dl.acm.org
Deep neural networks and machine-learning algorithms are pervasively used in several
applications, ranging from computer vision to computer security. In most of these …

Towards deep learning models resistant to adversarial attacks

A Madry, A Makelov, L Schmidt, D Tsipras… - arxiv preprint arxiv …, 2017‏ - arxiv.org
Recent work has demonstrated that deep neural networks are vulnerable to adversarial
examples---inputs that are almost indistinguishable from natural data and yet classified …

A closer look at memorization in deep networks

D Arpit, S Jastrzębski, N Ballas… - International …, 2017‏ - proceedings.mlr.press
We examine the role of memorization in deep learning, drawing connections to capacity,
generalization, and adversarial robustness. While deep networks are capable of memorizing …

Why do adversarial attacks transfer? explaining transferability of evasion and poisoning attacks

A Demontis, M Melis, M Pintor, M Jagielski… - 28th USENIX security …, 2019‏ - usenix.org
Transferability captures the ability of an attack against a machine-learning model to be
effective against a different, potentially unknown, model. Empirical evidence for …

Gradient descent maximizes the margin of homogeneous neural networks

K Lyu, J Li - arxiv preprint arxiv:1906.05890, 2019‏ - arxiv.org
In this paper, we study the implicit regularization of the gradient descent algorithm in
homogeneous neural networks, including fully-connected and convolutional neural …

Generalization in deep learning

K Kawaguchi, LP Kaelbling, Y Bengio - arxiv preprint arxiv …, 2017‏ - cambridge.org
This chapter provides theoretical insights into why and how deep learning can generalize
well, despite its large capacity, complexity, possible algorithmic instability, non-robustness …