V
Valeriia Cherepanova
Researcher at University of Maryland, College Park
Publications - 14
Citations - 219
Valeriia Cherepanova is an academic researcher from University of Maryland, College Park. The author has contributed to research in topics: Facial recognition system & Computer science. The author has an hindex of 6, co-authored 12 publications receiving 84 citations. Previous affiliations of Valeriia Cherepanova include University College London.
Papers
More filters
Posted Content
Strong Data Augmentation Sanitizes Poisoning and Backdoor Attacks Without an Accuracy Tradeoff
Eitan Borgnia,Valeriia Cherepanova,Liam Fowl,Amin Ghiasi,Jonas Geiping,Micah Goldblum,Tom Goldstein,Arjun Gupta +7 more
TL;DR: It is found that strong data augmentations, such as mixup and CutMix, can significantly diminish the threat of poisoning and backdoor attacks without trading off performance.
Proceedings Article
Unraveling Meta-Learning: Understanding Feature Representations for Few-Shot Tasks
TL;DR: A better understanding of the underlying mechanics of meta-learning is developed and a regularizer is developed which boosts the performance of standard training routines for few-shot classification.
Journal ArticleDOI
Deep learning of HIV field-based rapid tests.
Valérian Turbé,Carina Herbst,Thobeka Mngomezulu,Sepehr Meshkinfamfard,Nondumiso Dlamini,Thembani Mhlongo,Theresa Smit,Valeriia Cherepanova,Koki Shimada,Jobie Budd,Jobie Budd,Nestor Arsenov,Steven G. Gray,Deenan Pillay,Kobus Herbst,Maryam Shahmanesh,Rachel A. McKendry,Rachel A. McKendry +17 more
TL;DR: In this article, the authors used deep learning to classify images of rapid human immunodeficiency virus (HIV) tests acquired in rural South Africa using newly developed image capture protocols with the Samsung SM-P585 tablet.
Journal ArticleDOI
Development of PancRISK, a urine biomarker-based risk score for stratified screening of pancreatic cancer patients.
Oleg Blyuss,Oleg Blyuss,Oleg Blyuss,Alexey Zaikin,Alexey Zaikin,Valeriia Cherepanova,Daniel Munblit,Daniel Munblit,Elena M. Kiseleva,Olga M. Prytomanova,Stephen W. Duffy,Tatjana Crnogorac-Jurcevic +11 more
TL;DR: PancRISK score enables easy interpretation of the biomarker panel data and is currently being tested to confirm that it can be used for stratification of patients at risk of developing pancreatic cancer completely non-invasively, using urine samples.
Proceedings ArticleDOI
Strong Data Augmentation Sanitizes Poisoning and Backdoor Attacks Without an Accuracy Tradeoff
Eitan Borgnia,Valeriia Cherepanova,Liam Fowl,Amin Ghiasi,Jonas Geiping,Micah Goldblum,Tom Goldstein,Arjun Gupta +7 more
TL;DR: In this paper, strong data augmentations, such as mixup and CutMix, can significantly diminish the threat of poisoning and backdoor attacks without trading off performance, and they further verify the effectiveness of this simple defense against adaptive poisoning methods, and compare to baselines including the popular differentially private SGD (DP-SGD) defense.