TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
The TensorFlow interface and an implementation of that interface that is built at Google are described, which has been used for conducting research and for deploying machine learning systems into production across more than a dozen areas of computer science and other fields.
SmoothGrad: removing noise by adding noise
SmoothGrad is introduced, a simple method that can help visually sharpen gradient-based sensitivity maps and lessons in the visualization of these maps are discussed.
Google’s Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation
- Melvin Johnson, M. Schuster, J. Dean
- Computer ScienceInternational Conference on Topology, Algebra and…
- 14 November 2016
This work proposes a simple solution to use a single Neural Machine Translation (NMT) model to translate between multiple languages using a shared wordpiece vocabulary, and introduces an artificial token at the beginning of the input sentence to specify the required target language.
Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)
- Been Kim, M. Wattenberg, R. Sayres
- Computer ScienceInternational Conference on Machine Learning
- 30 November 2017
Concept Activation Vectors (CAVs) are introduced, which provide an interpretation of a neural net's internal state in terms of human-friendly concepts, and may be used to explore hypotheses and generate insights for a standard image classification network as well as a medical application.
A fuzzy commitment scheme
Because the fuzzy commitment scheme is tolerant of error, it is capable of protecting biometric data just as conventional cryptographic techniques, like hash functions, are used to protect alphanumeric passwords.
ManyEyes: a Site for Visualization at Internet Scale
- F. Viégas, M. Wattenberg, F. V. Ham, J. Kriss, Matt McKeon
- Computer Science, ArtIEEE Transactions on Visualization and Computer…
- 1 November 2007
The design and deployment of Many Eyes is described, a public Web site where users may upload data, create interactive visualizations, and carry on discussions to support collaboration around visualizations at a large scale by fostering a social style of data analysis.
Studying cooperation and conflict between authors with history flow visualizations
- F. Viégas, M. Wattenberg, Kushal Dave
- Computer ScienceInternational Conference on Human Factors in…
- 25 April 2004
This paper investigates the dynamics of Wikipedia, a prominent, thriving wiki, and focuses on the relevance of authorship, the value of community surveillance in ameliorating antisocial behavior, and how authors with competing perspectives negotiate their differences.
Ad click prediction: a view from the trenches
- H. B. McMahan, Gary Holt, J. Kubica
- Computer ScienceKnowledge Discovery and Data Mining
- 11 August 2013
The goal of this paper is to highlight the close relationship between theoretical advances and practical engineering in this industrial setting, and to show the depth of challenges that appear when applying traditional machine learning methods in a complex dynamic system.
Ordered and quantum treemaps: Making effective use of 2D space to display hierarchies
A new "strip" treemap algorithm is introduced which addresses shortcomings of previous algorithms, and is shown to be more stable, while maintaining relatively favorable aspect ratios of the constituent rectangles.
Talk Before You Type: Coordination in Wikipedia
- F. Viégas, M. Wattenberg, J. Kriss, F. V. Ham
- SociologyHawaii International Conference on System…
- 3 January 2007
The results suggest that despite the potential for anarchy, the Wikipedia community places a strong emphasis on group coordination, policy, and process.