H
Hana Ajakan
Researcher at Laval University
Publications - 5
Citations - 6851
Hana Ajakan is an academic researcher from Laval University. The author has contributed to research in topics: Domain (software engineering) & Artificial neural network. The author has an hindex of 5, co-authored 5 publications receiving 5084 citations.
Papers
More filters
Book ChapterDOI
Domain-adversarial training of neural networks
Yaroslav Ganin,Evgeniya Ustinova,Hana Ajakan,Pascal Germain,Hugo Larochelle,François Laviolette,Mario Marchand,Victor Lempitsky +7 more
TL;DR: In this article, a new representation learning approach for domain adaptation is proposed, in which data at training and test time come from similar but different distributions, and features that cannot discriminate between the training (source) and test (target) domains are used to promote the emergence of features that are discriminative for the main learning task on the source domain.
Domain-Adversarial Training of Neural Networks.
Yaroslav Ganin,Evgeniya Ustinova,Hana Ajakan,Pascal Germain,Hugo Larochelle,François Laviolette,Mario Marchand,Victor Lempitsky +7 more
TL;DR: A new representation learning approach for domain adaptation, in which data at training and test time come from similar but different distributions, which can be achieved in almost any feed-forward model by augmenting it with few standard layers and a new gradient reversal layer.
Posted Content
Domain-Adversarial Neural Networks
TL;DR: A new neural network learning algorithm suited to the context of domain adaptation, in which data at training and test time come from similar but different distributions, which has better performance than either a standard neural networks and a SVM.
Posted Content
Domain-Adversarial Training of Neural Networks
Yaroslav Ganin,Evgeniya Ustinova,Hana Ajakan,Pascal Germain,Hugo Larochelle,François Laviolette,Mario Marchand,Victor Lempitsky +7 more
TL;DR: In this article, a new representation learning approach for domain adaptation is proposed, in which data at training and test time come from similar but different distributions, and features that cannot discriminate between the training (source) and test (target) domains are used to promote the emergence of features that are discriminative for the main learning task on the source domain.
Book ChapterDOI
Andrana: Quick and Accurate Malware Detection for Android
Andrew Bedford,Sébastien Garvin,Josée Desharnais,Nadia Tawbi,Hana Ajakan,Frédéric Audet,Bernard Lebel +6 more
TL;DR: This paper presents Andrana, a lightweight malware detection tool for Android that leverages machine learning techniques and static analysis to determine, with an accuracy of 94.90%, if an application is malicious.