O
Oriol Vinyals
Researcher at Google
Publications - 218
Citations - 121048
Oriol Vinyals is an academic researcher from Google. The author has contributed to research in topics: Artificial neural network & Reinforcement learning. The author has an hindex of 84, co-authored 200 publications receiving 82365 citations. Previous affiliations of Oriol Vinyals include University of California, San Diego & University of California, Berkeley.
Papers
More filters
Posted Content
An Online Sequence-to-Sequence Model Using Partial Conditioning
TL;DR: In this article, an encoder recurrent neural network (RNN) is used to compute features at the same frame rate as the input, and a transducer RNN that operates over blocks of input steps.
Patent
Using Hierarchical Representations for Neural Network Architecture Searching
TL;DR: In this paper, a computer-implemented method for automatically determining a neural network architecture is presented, in which the modification is performed by selecting a level, selecting two nodes at that level, and modifying, removing or adding an edge between those nodes according to operations associated with lower levels of the hierarchy.
Posted Content
WikiGraphs: A Wikipedia Text - Knowledge Graph Paired Dataset
TL;DR: The WikiGraphs dataset as mentioned in this paper pairs Wikipedia articles with a knowledge graph to facilitate the research in conditional text generation, graph generation and graph representation learning, which is a dataset of Wikipedia articles paired with knowledge graphs.
HiP: Hierarchical Perceiver
Joao Carreira,Skanda Koppula,Deb Zoran,Adrià Recasens,Catalin Ionescu,Olivier J. Hénaff,Evan Shelhamer,Relja Arandjelovic,Matthew Botvinick,Oriol Vinyals,Karen Simonyan,A. Zisserman,Andrew Jaegle +12 more
TL;DR: In this article , Hierarchical Perceiver (HiP) is proposed to learn dense low-dimensional positional embeddings for high-resolution images and videos, which can handle up to a few hundred thousand inputs.
Posted Content
Why Size Matters: Feature Coding as Nystrom Sampling
TL;DR: A novel view of feature extraction pipelines that rely on a coding step followed by a linear classifier based on kernel methods and Nystrom sampling is proposed, which may help explaining the positive effect of the codebook size and justifying the need to stack more layers, as flat models empirically saturate as the authors add more complexity.