A
Andrei Rusu
Researcher at West University of Timișoara
Publications - 64
Citations - 34789
Andrei Rusu is an academic researcher from West University of Timișoara. The author has contributed to research in topics: Reinforcement learning & Meta learning (computer science). The author has an hindex of 25, co-authored 61 publications receiving 23267 citations. Previous affiliations of Andrei Rusu include Technical University of Cluj-Napoca & Ovidius University.
Papers
More filters
Proceedings ArticleDOI
Reinforcement and Imitation Learning for Diverse Visuomotor Skills
Yuke Zhu,Ziyu Wang,Josh Merel,Andrei Rusu,Tom Erez,Serkan Cabi,Saran Tunyasuvunakool,János Kramár,Raia Hadsell,Nando de Freitas,Nicolas Heess +10 more
TL;DR: This article propose a model-free deep reinforcement learning method that leverages a small amount of demonstration data to assist a reinforcement learning agent for robotic manipulation tasks and train end-to-end visuomotor policies that map directly from RGB camera inputs to joint velocities.
Journal ArticleDOI
Embracing Change: Continual Learning in Deep Neural Networks
TL;DR: This review relates continual learning to the learning dynamics of neural networks, highlighting the potential it has to considerably improve data efficiency and consider the many new biologically inspired approaches that have emerged in recent years.
Journal ArticleDOI
Imagine All the People: How the Brain Creates and Uses Personality Models to Predict Behavior
Demis Hassabis,R. Nathan Spreng,Andrei Rusu,Clifford A. Robbins,Raymond A. Mar,Daniel L. Schacter +5 more
TL;DR: The results suggest that distinct regions code for personality traits, and that the brain combines these traits to represent individuals, and then uses this "personality model" to predict the behavior of others in novel situations.
Proceedings Article
Continual Unsupervised Representation Learning.
TL;DR: The proposed approach (CURL) performs task inference directly within the model, is able to dynamically expand to capture new concepts over its lifetime, and incorporates additional rehearsal-based techniques to deal with catastrophic forgetting.
Posted Content
Policy Distillation
Andrei Rusu,Sergio Gomez Colmenarejo,Caglar Gulcehre,Guillaume Desjardins,James Kirkpatrick,Razvan Pascanu,Volodymyr Mnih,Koray Kavukcuoglu,Raia Hadsell +8 more
TL;DR: A novel method called policy distillation is presented that can be used to extract the policy of a reinforcement learning agent and train a new network that performs at the expert level while being dramatically smaller and more efficient.