Muhammad Umer
Rowan University
13 Papers
23 Citations
Muhammad Umer is an academic researcher from Rowan University. The author has contributed to research in topics: Computer science & Backdoor. The author has an hindex of 6, co-authored 13 publications. Previous affiliations of Muhammad Umer include College of Electrical and Mechanical Engineering.
Chat about Author
Papers
Machine Learning Analysis of Digital Clock Drawing Test Performance for Differential Classification of Mild Cognitive Impairment Subtypes Versus Alzheimer's Disease
Russell Binaco,Nicholas Calzaretto,Jacob R. Epifano,Sean McGuire,Muhammad Umer,Sheina Emrani,Victor Wasserman,David J. Libon,Robi Polikar +8 more
TL;DR: Applying machine learning to standard neuropsychological tests promises to be an effective first line screening method for classification of non-MCI and MCI subtypes and early identification of emergent neurodegenerative illness is criterial for better disease management.
70
Electrocardiogram Feature Extraction and Pattern Recognition Using a Novel Windowing Algorithm
Muhammad Umer,Bilal Ahmed Bhatti,Muhammad Hammad Tariq,Muhammad Zia-ul-Hassan,Muhammad Yaqub Khan,Tahir Zaidi +5 more
TL;DR: A simple and efficient way of detecting ECG features that are P, Q, R, S and T waves is presented that has been tested on ECG simulator data and also on different records of the MIT-BIH arrhythmia database, producing satisfactory results.
Learning under extreme verification latency quickly: FAST COMPOSE
Muhammad Umer,Christopher Frederickson,Robi Polikar +2 more
- 01 Dec 2016
TL;DR: A modification to COMPOSE (COMPacted Object Sample Extraction) is described that allows the algorithm to work without core support extraction, and is called FAST COMPOSE, both in accuracy and in execution time.
19
•Posted Content
Targeted Forgetting and False Memory Formation in Continual Learners through Adversarial Backdoor Attacks
TL;DR: This effort explores the vulnerability of Elastic Weight Consolidation and shows that an intelligent adversary can take advantage of EWC’s continual learning capabilities to cause gradual and deliberate forgetting by introducing small amounts of misinformation to the model during training.
15
Vulnerability of Covariate Shift Adaptation Against Malicious Poisoning Attacks
Muhammad Umer,Christopher Frederickson,Robi Polikar +2 more
- 14 Jul 2019
TL;DR: The vulnerability of unconstrained least squares importance fitting (uLSIF), an algorithm used for computing the importance ratio for covariate shift domain adaptation problems, is explored and it is demonstrated that importance ratio estimation through uLSIF algorithm can be easily compromised.
11