TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
By Martín Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Z. Chen, C. Citro, Greg S. Corrado, Andy Davis, Joseph Dean, Matthieu Devin and 31 more. 2016
11,798 citations recorded by the source corpus.
TensorFlow is an interface for expressing machine learning algorithms, and an implementation for executing such algorithms. A computation expressed using TensorFlow can be executed with little or no change on a wide variety of heterogeneous systems, ranging from mobile devices such as phones and tablets up to large-scale distributed systems of hundreds of machines and thousands of computational devices such as GPU cards. The system is flexible and can be used to express a wide variety of algorithms, including training and inference algorithms for deep neural network models, and it 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, including speech recognition, computer vision, robotics, information retrieval, natural language processing, geographic information extraction, and computational drug discovery. This paper describes the TensorFlow interface and an implementation of that interface that we have built at Google. The TensorFlow API and a reference implementation were released as an open-source package under the Apache 2.0 license in November, 2015 and are available at www.tensorflow.org.
Subfields: Hardware and Architecture, Information Systems, Information Systems and Management, Optimization & Training
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Hardware and Architecture
- Matthias Hein
- Nicholas Carlini
- Yoshua Bengio
- Bernt Schiele
- Ian J. Goodfellow
- Bo Li
- Wieland Brendel
- Nicolas Papernot
In Information Systems
- Xiangnan He
- Miltiadis Allamanis
- Percy Liang
- Yoshua Bengio
- Luke S. Zettlemoyer
- Charles Sutton
- Jianfeng Gao
- Marc Brockschmidt
In Information Systems and Management
- Jung-Woo Ha
- Ahmed Hassan Awadallah
- Brandon Amos
- J. Kolter
- Lei Xu
- Sebastian Riedel
- Miltiadis Allamanis
- Quoc V. Le
In Optimization & Training
- Yoshua Bengio
- Peter Richtárik
- Sergey Levine
- Martin Jaggi
- Quoc V. Le
- Simon Lacoste-Julien
- Sebastian U. Stich
- Jakub Konecný
Citations in the corpus
- Can Active Memory Replace Attention? (2016, positive): “Better 2Our model was implemented using TensorFlow [26].”
- Inductive Representation Learning on Large Graphs (2017, neutral): “All models were implemented in TensorFlow [1] with the Adam optimizer [16] (except DeepWalk, which performed better with the vanilla gradient descent optimizer).”
- RL$^2$: Fast Reinforcement Learning via Slow Reinforcement Learning (2016, methodology): “All experiments are implemented using TensorFlow (Abadi et al., 2016) and rllab (Duan et al., 2016).”
- Array programming with NumPy (2020, methodology): “PyTorch6 and Tensorflow7 provide tensor APIs with NumPy-inspired semantics.”
- Parallel WaveNet: Fast High-Fidelity Speech Synthesis (2017, methodology): “Both models were implemented in Tensorflow [1] and compiled with XLA.”
- Conditional Image Generation with PixelCNN Decoders (2016, methodology): “We used 200 K synchronous updates over 32 GPUs in TensorFlow [1]”
- Density estimation using Real NVP (2016, background): “The authors thank the developers of Tensorflow [1].”
- Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering (2016, methodology): “Moreover, this method avoids the Fourier basis altogether, thus the expensive eigenvalue decomposition (EVD) necessary to compute it as well as the need to store the basis, a matrix of size n2.”
- PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation (2016, background): “We also show the effects of our network’s hyperparameters.”