Jarrod Hollis

Jarrod Hollis

Fourth year CS Ph.D. student, with a focus on developing countermeasures for cost-aware adversarial machine learning.

Publications

Identifiable Solutions to Foreground Signature Extraction from Hyperspectral Images in an Intimate Mixing Scenario

Published in arXiv preprint, 2023

This paper extends our ICASSP 2020 conference paper, including formal identifiability analysis of the extraction problem and additional algorithms and experiments.

Recommended citation: J. Hollis, R. Raich, J. Kim, B. Fishbain and S. Kendler, "Identifiable Solutions to Foreground Signature Extraction from Hyperspectral Images in an Intimate Mixing Scenario," arXiv preprint, 2023, arXiv:2303.11479. https://arxiv.org/abs/2303.11479

Adversarial learning via probabilistic proximity analysis

Published in IEEE International Conference on Acoustics, Speech and Signal Processing, 2021

This paper proposes the Lead-Class KNN classifier as a defense against high-cost adversarial attacks.

Recommended citation: J. Hollis, J. Kim and R. Raich, "Adversarial Learning via Probabilistic Proximity Analysis," ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Toronto, ON, Canada, 2021, pp. 3830-3834, doi: 10.1109/ICASSP39728.2021.9414096. https://ieeexplore.ieee.org/abstract/document/9414096/

Foreground signature extraction for an intimate mixing model in hyperspectral image classification

Published in IEEE International Conference on Acoustics, Speech and Signal Processing, 2020

This paper explores a framework for foreground material signature extraction from hyperspectral images based on a proposed patch model.

Recommended citation: J. Hollis, R. Raich, J. Kim, B. Fishbain and S. Kendler, "Foreground Signature Extraction for an Intimate Mixing Model in Hyperspectral Image Classification," ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain, 2020, pp. 4732-4736, doi: 10.1109/ICASSP40776.2020.9053456. https://ieeexplore.ieee.org/abstract/document/9053456