Výsledky vyhledávání - "Manville, Keith"

Report

Machine Learning Model Attribution Challenge

Autor: Merkhofer, Elizabeth, Chaudhari, Deepesh, Anderson, Hyrum S., Manville, Keith, Wong, Lily, Gante, João

We present the findings of the Machine Learning Model Attribution Challenge. Fine-tuned machine learning models may derive from other trained models without obvious attribution characteristics. In this challenge, participants identify the publicly-av

Externí odkaz: http://arxiv.org/abs/2302.06716

Zobrazit plný text záznamu

Report

Adversarial Attack Attribution: Discovering Attributable Signals in Adversarial ML Attacks

Autor: Dotter, Marissa, Xie, Sherry, Manville, Keith, Harguess, Josh, Busho, Colin, Rodriguez, Mikel

Machine Learning (ML) models are known to be vulnerable to adversarial inputs and researchers have demonstrated that even production systems, such as self-driving cars and ML-as-a-service offerings, are susceptible. These systems represent a target f

Externí odkaz: http://arxiv.org/abs/2101.02899

Zobrazit plný text záznamu

Report

APRICOT: A Dataset of Physical Adversarial Attacks on Object Detection

Autor: Braunegg, Anneliese, Chakraborty, Amartya, Krumdick, Michael, Lape, Nicole, Leary, Sara, Manville, Keith, Merkhofer, Elizabeth, Strickhart, Laura, Walmer, Matthew

Physical adversarial attacks threaten to fool object detection systems, but reproducible research on the real-world effectiveness of physical patches and how to defend against them requires a publicly available benchmark dataset. We present APRICOT,

Externí odkaz: http://arxiv.org/abs/1912.08166

Zobrazit plný text záznamu

Report

Overhead Detection: Beyond 8-bits and RGB

Autor: Mace, Eliza, Manville, Keith, Barbu-McInnis, Monica, Laielli, Michael, Klaric, Matthew, Dooley, Samuel

Publikováno v: Naval Applications of Machine Learning, February 13, 2018

This study uses the challenging and publicly available SpaceNet dataset to establish a performance baseline for a state-of-the-art object detector in satellite imagery. Specifically, we examine how various features of the data affect building detecti

Externí odkaz: http://arxiv.org/abs/1808.02443

Zobrazit plný text záznamu