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
Tackling AlfWorld with Action Attention and Common Sense from Pretrained LMs
TL;DR: A novel question answering framework to simplify observations and an agent that handles arbitrary roll-out length and action space size based on action based on attention are implemented.
Posted Content
Accelerating Robotic Reinforcement Learning via Parameterized Action Primitives
TL;DR: In this paper, the authors propose a library of robot action primitives (RAPS), parameterized with arguments that are learned by an RL policy, which can be transferred across robots, tasks and environments.
Zero-shot Transfer Learning on Heterogeneous Graphs via Knowledge Transfer Networks
TL;DR: A zero-shot transfer learning module for HGNNs called a Knowledge Transfer Network (KTN) is proposed that transfers knowledge from label-abundant node types to zero-labeled node types through rich relational information given in the HG.
Posted Content
Close Category Generalization.
TL;DR: It is found that performing well on close category generalization correlates with learning a good representation of an unseen class and with finding a good initialization for few-shot learning.
Posted Content
Mismatched No More: Joint Model-Policy Optimization for Model-Based RL.
TL;DR: In this paper, a single objective for jointly training the model and the policy, such that updates to either component increases a lower bound on expected return is proposed, which mends the objective mismatch in prior work.