R
Ruslan Salakhutdinov
Researcher at Carnegie Mellon University
Publications - 457
Citations - 142495
Ruslan Salakhutdinov is an academic researcher from Carnegie Mellon University. The author has contributed to research in topics: Computer science & Artificial neural network. The author has an hindex of 107, co-authored 410 publications receiving 115921 citations. Previous affiliations of Ruslan Salakhutdinov include Carnegie Learning & University of Toronto.
Papers
More filters
Graph Adversarial Networks: Protecting Information against Adversarial Attacks
Peiyuan Liao,Han Zhao,Keyulu Xu,Tommi S. Jaakkola,Geoff Gordon,Stefanie Jegelka,Ruslan Salakhutdinov +6 more
TL;DR: A minimax game between the desired GNN encoder and the worst-case attacker is proposed, and the resulting adversarial training creates a strong defense against inference attacks, while only suffering small loss in task performance.
Posted Content
On Emergent Communication in Competitive Multi-Agent Teams
TL;DR: The results show that an external competitive influence leads to improved accuracy and generalization, as well as faster emergence of communicative languages that are more informative and compositional.
Proceedings ArticleDOI
Imitating Past Successes can be Very Suboptimal
TL;DR: It is proved that existing outcome-conditioned imitation learning methods do not necessarily improve the policy; rather, in some settings they can decrease the expected reward.
Posted Content
Exploring Controllable Text Generation Techniques
TL;DR: In this paper, the authors provide a new schema of the pipeline of the generation process by classifying it into five modules and present an overview of different techniques used to perform the modulation of these modules.
Proceedings Article
Weakly-Supervised Reinforcement Learning for Controllable Behavior
TL;DR: In this article, a framework for using weak supervision to automatically disentangle the semantically meaningful subspace of tasks from the enormous space of nonsensical "chaff" tasks is introduced.