J
Jonathon Shlens
Researcher at Google
Publications - 116
Citations - 88633
Jonathon Shlens is an academic researcher from Google. The author has contributed to research in topics: Object detection & Artificial neural network. The author has an hindex of 53, co-authored 116 publications receiving 63492 citations. Previous affiliations of Jonathon Shlens include Salk Institute for Biological Studies.
Papers
More filters
Fast, Accurate Detection of 100,000 Object Classes on a Single Machine: Technical Supplement
TL;DR: In this article, the deformable part models (DPMs) are trained on HOG-based part filters and, during inference, counts of hashing collisions summed over all hash bands serve as a proxy for part-filter / sliding window dot products, i.e., filter responses.
Patent
Estimating rate of change of documents
TL;DR: In this paper, a first document from a corpus and metadata for the first document were obtained, and the change rates for the second documents were estimated based on the metadata and the first documents' change rates.
Patent
Neural architecture search using a performance prediction neural network
Wei Hua,Barret Zoph,Jonathon Shlens,Chenxi Liu,Jonathan Huang,Li Jia,Li Fei-Fei,Kevin Murphy +7 more
TL;DR: In this article, a method for determining an architecture for a task neural network configured to perform a particular machine learning task is described, which includes obtaining data specifying a current set of candidate architectures for the task neural networks; for each candidate architecture in the current set, processing the data specifying the candidate architecture using a performance prediction neural network having multiple performance prediction parameters, the performance prediction Neural Network being configured to process the data specified the candidate architectures in accordance with current values of the performance predictions parameters.
Posted Content
Accelerating Training of Deep Neural Networks with a Standardization Loss
TL;DR: A standardization loss is proposed to replace existing normalization methods with a simple, secondary objective loss that accelerates training on both small- and large-scale image classification experiments, works with a variety of architectures, and is largely robust to training across different batch sizes.
Posted Content
Scalable Scene Flow from Point Clouds in the Real World.
TL;DR: In this article, a large-scale dataset for scene flow estimation derived from corresponding tracked 3D objects is introduced, called FastFlow3D, which is $1,000,000$\times$ larger than previous real-world datasets in terms of the number of annotated frames.