F
Fillia Makedon
Researcher at University of Texas at Arlington
Publications - 398
Citations - 6463
Fillia Makedon is an academic researcher from University of Texas at Arlington. The author has contributed to research in topics: Wireless sensor network & Robot. The author has an hindex of 37, co-authored 385 publications receiving 5100 citations. Previous affiliations of Fillia Makedon include Dartmouth College & University of the Aegean.
Papers
More filters
Posted Content
A Survey on Contrastive Self-supervised Learning
TL;DR: This paper provides an extensive review of self-supervised methods that follow the contrastive approach, explaining commonly used pretext tasks in a contrastive learning setup, followed by different architectures that have been proposed so far.
Proceedings Article
Learning from incomplete ratings using non-negative matrix factorization
TL;DR: Two variations on Non-negative Matrix Factorization (NMF) are introduced: one based on the Expectation-Maximization (EM) procedure and the other a Weighted Nonnegative Matrix factorization (WNMF), which obtain the best prediction performance compared with other popular collaborative filtering algorithms in experiments.
Journal ArticleDOI
A Survey on Contrastive Self-Supervised Learning
TL;DR: In contrastive self-supervised learning as discussed by the authors, augmented versions of the same sample close to each other while trying to push away embeddings from different samples is used to learn representations for several downstream tasks.
Proceedings ArticleDOI
Fast approximation algorithms for multicommodity flow problems
TL;DR: It is proved that a (simple) k-commodity flow problem can be approximately solved by approximately solving O(k log2n) single-comodity minimum-cost flow problems, and the first polynomial-time combinatorial algorithms for approximately solving the multicommodation flow problem are described.
Journal ArticleDOI
HykGene: a hybrid approach for selecting marker genes for phenotype classification using microarray gene expression data
TL;DR: A novel hybrid approach that combines gene ranking and clustering analysis is developed that is capable of selecting relatively few marker genes while offering the same or better leave-one-out cross-validation accuracy compared with approaches that use top-ranked genes directly for classification.