TensorFlow: a system for large-scale machine learning

M. Abadi, A. Agarwal, P. Barham, E. Brevdo, Z. Chen, C. Citro, G. S. Corrado, A. Davis, J. Dean, M. Devin, S. Ghemawat, I. J. Goodfellow, A. Harp, G. Irving, M. Isard, Y. Jia, R. Józefowicz, L. Kaiser, M. Kudlur, J. Levenberg, D. Mane, R. Monga, S. Moore, D. G. Murray, C. Olah, M. Schuster, J. Shlens, B. Steiner, I. Sutskever, K. Talwar, P. A. Tucker, V. Vanhoucke, V. Vasudevan, F. B. Viégas, O. Vinyals, P. Warden, M. Wattenberg, M. Wicke, Y. Yu, and X. Zheng. TensorFlow: Large-scale machine learning on heterogeneous distributed systems. arXiv preprint, 1603.04467, 2016. arxiv.org/abs/1603.04467. Software available from tensorflow.org.

R. Al-Rfou, G. Alain, A. Almahairi, C. Angermueller, D. Bahdanau, N. Ballas, F. Bastien, J. Bayer, A. Belikov, A. Belopolsky, Y. Bengio, A. Bergeron, J. Bergstra, V. Bisson, J. Bleecher Snyder, N. Bouchard, N. Boulanger-Lewandowski, X. Bouthillier, A. de Brébisson, O. Breuleux, P.- L. Carrier, K. Cho, J. Chorowski, P. Christiano, T. Cooijmans, M.-A. Côté, M. Côté, A. Courville, Y. N. Dauphin, O. Delalleau, J. Demouth, G. Desjardins, S. Dieleman, L. Dinh, M. Ducoffe, V. Dumoulin, S. Ebrahimi Kahou, D. Erhan, Z. Fan, O. Firat, M. Germain, X. Glorot, I. Goodfellow, M. Graham, C. Gulcehre, P. Hamel, I. Harlouchet, J.-P. Heng, B. Hidasi, S. Honari, A. Jain, S. Jean, K. Jia, M. Korobov, V. Kulkarni, A. Lamb, P. Lamblin, E. Larsen, C. Laurent, S. Lee, S. Lefrancois, S. Lemieux, N. Léonard, Z. Lin, J. A. Livezey, C. Lorenz, J. Lowin, Q. Ma, P.-A. Manzagol, O. Mastropietro, R. T. McGibbon, R. Memisevic, B. van Merriënboer, V. Michalski, M. Mirza, A. Orlandi, C. Pal, R. Pascanu, M. Pezeshki, C. Raffel, D. Renshaw, M. Rocklin, A. Romero, M. Roth, P. Sadowski, J. Salvatier, F. Savard, J. Schlüter, J. Schulman, G. Schwartz, I. V. Serban, D. Serdyuk, S. Shabanian, E. Simon, S. Spieckermann, S. R. Subramanyam, J. Sygnowski, J. Tanguay, G. van Tulder, J. Turian, S. Urban, P. Vincent, F. Visin, H. de Vries, D. Warde-Farley, D. J. Webb, M. Willson, K. Xu, L. Xue, L. Yao, S. Zhang, and Y. Zhang. Theano: A Python framework for fast computation of mathematical expressions. arXiv preprint, 1605.02688, 2016. arxiv.org/abs/1605.02688.

A. Angelova, A. Krizhevsky, and V. Vanhoucke. Pedestrian detection with a large-field-of-view deep network. In Proceedings of ICRA, pages 704-711. IEEE, 2015. www.vision.caltech.edu/anelia/publications/Angelova15LFOV.pdf.

Arvind and D. E. Culler. Dataflow architectures. In Annual Review of Computer Science Vol. 1, 1986, pages 225-253. Annual Reviews Inc., 1986. www.dtic.mil/cgi-bin/GetTRDoc?Location=U2&doc=GetTRDoc.pdf&AD=ADA166235.

J. Ba, V. Mnih, and K. Kavukcuoglu. Multiple object recognition with visual attention. arXiv preprint, 1412.7755, 2014. arxiv.org/abs/1412.7755.

Y. Bengio, R. Ducharme, P. Vincent, and C. Jauvin. A neural probabilistic language model. Journal of Machine Learning Research, 3:1137-1155, 2003. jmlr.org/papers/volume3/bengio03a/bengio03a.pdf.

T. Brants and A. Franz. Web 1T 5-gram version 1, 2006. catalog.ldc.upenn.edu/LDC2006T13.

R. H. Byrd, G. M. Chin, J. Nocedal, and Y. Wu. Sample size selection in optimization methods for machine learning. Mathematical Programming, 134(1):127-155, 2012. dx.doi.org/10.1007/s10107-012-0572-5.

C. Chelba, T. Mikolov, M. Schuster, Q. Ge, T. Brants, and P. Koehn. One billion word benchmark for measuring progress in statistical language modeling. arXiv preprint, 1312.3005, 2013. arxiv.org/abs/1312.3005.

J. Chen, R. Monga, S. Bengio, and R. Jozefowicz. Revisiting distributed synchronous SGD. In Proceedings of ICLR Workshop Track, 2016. arxiv.org/abs/1604.00981.

T. Chen, M. Li, Y. Li, M. Lin, N. Wang, M. Wang, T. Xiao, B. Xu, C. Zhang, and Z. Zhang. MXNet: A flexible and efficient machine learning library for heterogeneous distributed systems. In Proceedings of LearningSys, 2015. www.cs.cmu.edu/~muli/file/mxnet-learning-sys.pdf.

H.-T. Cheng, L. Koc, J. Harmsen, T. Shaked, T. Chandra, H. Aradhye, G. Anderson, G. Corrado, W. Chai, M. Ispir, R. Anil, Z. Haque, L. Hong, V. Jain, X. Liu, and H. Shah. Wide & deep learning for recommender systems. arXiv preprint, 1606.07792, 2016. arxiv.org/abs/1606.07792.

S. Chetlur, C. Woolley, P. Vandermersch, J. Cohen, J. Tran, B. Catanzaro, and E. Shelhamer. cuDNN: Efficient primitives for deep learning. arXiv preprint, 1410.0759, 2014. arxiv.org/abs/1410.0759.

T. Chilimbi, Y. Suzue, J. Apacible, and K. Kalyanaraman. Project Adam: Building an efficient and scalable deep learning training system. In Proceedings of OSDI, pages 571-582, 2014. www.usenix.org/system/files/conference/osdi14/osdi14-paper-chilimbi.pdf.

S. Chintala. convnet-benchmarks, 2016. github.com/soumith/convnet-benchmarks.

E. S. Chung, J. D. Davis, and J. Lee. LINQits: Big data on little clients. In Proceedings of ISCA, pages 261-272, 2013. www.microsoft.com/enus/research/wp-content/uploads/2013/06/ISCA13_-linqits.pdf.

R. Collobert, S. Bengio, and J. Mariéthoz. Torch: A modular machine learning software library. Technical report, IDIAP, 2002. infoscience.epfl.ch/record/82802/files/rr02-46.pdf.

H. Cui, H. Zhang, G. R. Ganger, P. B. Gibbons, and E. P. Xing. GeePS: Scalable deep learning on distributed GPUs with a GPU-specialized parameter server. In Proceedings of EuroSys, 2016. www.pdl.cmu.edu/PDL-FTP/CloudComputing/GeePS-cui-eurosys16.pdf.

A. Dai, C. Olah, and Q. V. Le. Document embedding with paragraph vectors. arXiv preprint, 1507.07998, 2015. arxiv.org/abs/1507.07998.

J. Dean, G. S. Corrado, R. Monga, K. Chen, M. Devin, Q. V. Le, M. Z. Mao, M. Ranzato, A. Senior, P. Tucker, K. Yang, and A. Y. Ng. Large scale distributed deep networks. In Proceedings of NIPS, pages 1232-1240, 2012. research.google.com/archive/large_deep_networks_nips2012.pdf.

J. Dean and S. Ghemawat. MapReduce: Simplified data processing on large clusters. In Proceedings of OSDI, pages 137-149, 2004. research.google.com/archive/mapreduceosdi04.pdf.

DMLC. MXNet for deep learning, 2016. github.com/dmlc/mxnet.

J. Duchi, E. Hazan, and Y. Singer. Adaptive subgradient methods for online learning and stochastic optimization. Journal of Machine Learning Research, 12:2121-2159, 2011. jmlr.org/papers/volume12/duchi11a/duchi11a.pdf.

A. Frome, G. S. Corrado, J. Shlens, S. Bengio, J. Dean, T. Mikolov, et al. DeVISE: A deep visual-semantic embedding model. In Proceedings of NIPS, pages 2121-2129, 2013. research.google.com/pubs/archive/41473.pdf.

J. Gonzalez-Dominguez, I. Lopez-Moreno, P. J. Moreno, and J. Gonzalez-Rodriguez. Frame-by-frame language identification in short utterances using deep neural networks. Neural Networks, 64:49-58, 2015. research.google.com/pubs/archive/42929.pdf.

I. J. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. C. Courville, and Y. Bengio. Generative adversarial nets. In Proceedings of NIPS, pages 2672- 2680, 2014. papers.nips.cc/paper/5423-generative-adversarial-nets.pdf.

Google Research. Tensorflow serving, 2016. tensorflow.github.io/serving/.

K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. In Proceedings of CVPR, pages 770-778, 2016. arxiv.org/abs/1512.03385.

G. Heigold, V. Vanhoucke, A. Senior, P. Nguyen, M. Ranzato, M. Devin, and J. Dean. Multilingual acoustic models using distributed deep neural networks. In Proceedings of ICASSP, pages 8619-8623, 2013. research.google.com/pubs/archive/40807.pdf.

G. E. Hinton. Learning distributed representations of concepts. In Proceedings of the Eighth Annual Conference of the Cognitive Science Society, pages 1-12, 1986. www.cogsci.ucsd.edu/~ajyu/Teaching/Cogs202_-sp13/Readings/hinton86.pdf.

G. E. Hinton, L. Deng, D. Yu, G. E. Dahl, A. Mohamed, N. Jaitly, A. Senior, V. Vanhoucke, P. Nguyen, T. N. Sainath, and B. Kingsbury. Deep neural networks for acoustic modeling in speech recognition: The shared views of four research groups. IEEE Signal Process. Mag., 29(6):82- 97, 2012. www.cs.toronto.edu/~gdahl/papers/deepSpeechReviewSPM2012.pdf.

S. Hochreiter and J. Schmidhuber. Long short-term memory. Neural computation, 9(8):1735-1780, 1997. deeplearning.cs.cmu.edu/pdfs/Hochreiter97_-lstm.pdf.

S. Ioffe and C. Szegedy. Batch normalization: Accelerating deep network training by reducing internal covariate shift. In Proceedings of ICML, pages 448-456, 2015. jmlr.org/proceedings/papers/v37/ioffe15.pdf.

M. Isard, M. Budiu, Y. Yu, A. Birrell, and D. Fetterly. Dryad: distributed data-parallel programs from sequential building blocks. In Proceedings of EuroSys, pages 59-72, 2007. www.microsoft.com/en-us/research/wpcontent/uploads/2007/03/eurosys07.pdf.

B. Jacob et al. gemmlowp: a small self-contained low-precision GEMM library, 2015. github.com/google/gemmlowp.

B. Jacob, G. Guennebaud, et al. Eigen library for linear algebra. eigen.tuxfamily.org.

S. Jean, K. Cho, R. Memisevic, and Y. Bengio. On using very large target vocabulary for neural machine translation. In Proceedings of ACL-ICJNLP, pages 1-10, July 2015. www.aclweb.org/anthology/P15-1001.

Y. Jia, E. Shelhamer, J. Donahue, S. Karayev, J. Long, R. Girshick, S. Guadarrama, and T. Darrell. Caffe: Convolutional architecture for fast feature embedding. In Proceedings of ACM Multimedia, pages 675-678, 2014. arxiv.org/abs/1408.5093.

M. I. Jordan. Serial order: A parallel distributed processing approach. ICS report 8608, Institute for Cognitive Science, UCSD, La Jolla, 1986. cseweb.ucsd.edu/~gary/PAPERSUGGESTIONS/Jordan-TR-8604.pdf.

N. Jouppi. Google supercharges machine learning tasks with TPU custom chip, 2016. cloudplatform.googleblog.com/2016/05/Google-supercharges-machine-learning-tasks-with-customchip.html.

R. Józefowicz, O. Vinyals, M. Schuster, N. Shazeer, and Y. Wu. Exploring the limits of language modeling. arXiv preprint, 1602.02410, 2016. arxiv.org/abs/1602.02410.

A. Karpathy, G. Toderici, S. Shetty, T. Leung, R. Sukthankar, and L. Fei-Fei. Large-scale video classification with convolutional neural networks. In Proceedings of CVPR, pages 1725-1732, 2014. research.google.com/pubs/archive/42455.pdf.

A. Krizhevsky. One weird trick for parallelizing convolutional neural networks. arXiv preprint, 1404.5997, 2014. arxiv.org/abs/1404.5997.

A. Krizhevsky, I. Sutskever, and G. E. Hinton. ImageNet classification with deep convolutional neural networks. In Proceedings of NIPS, pages 1106-1114, 2012. papers.nips.cc/paper/4824- imagenet-classification-with-deep-convolutional-neural-networks.pdf.

H. Larochelle, Y. Bengio, J. Louradour, and P. Lamblin. Exploring strategies for training deep neural networks. Journal of Machine Learning Research, 10:1-40, 2009. jmlr.org/papers/volume10/larochelle09a/larochelle09a.pdf.

A. Lavin and S. Gray. Fast algorithms for convolutional neural networks. In Proceedings of CVPR, pages 4013-4021, 2016. arxiv.org/abs/1509.09308.

Q. Le, M. Ranzato, R. Monga, M. Devin, G. Corrado, K. Chen, J. Dean, and A. Ng. Building high-level features using large scale unsupervised learning. In Proceedings of ICML, pages 81-88, 2012. research.google.com/archive/unsupervised_-icml2012.pdf.

Y. LeCun, C. Cortes, and C. J. Burges. The MNIST database of handwritten digits, 1998. yann.lecun.com/exdb/mnist/.

M. Li, D. G. Andersen, J. Park, A. J. Smola, A. Ahmed, V. Josifovski, J. Long, E. J. Shekita, and B.-Y. Su. Scaling distributed machine learning with the Parameter Server. In Proceedings of OSDI, pages 583-598, 2014. www.usenix.org/system/files/conference/osdi14/osdi14-paper-li_mu.pdf.

C. J. Maddison, A. Huang, I. Sutskever, and D. Silver. Move evaluation in Go using deep convolutional neural networks. arXiv preprint, 1412.6564, 2014. arxiv.org/abs/1412.6564.

F. McSherry, M. Isard, and D. G. Murray. Scalability! But at what COST? In Proceedings of HotOS, HOTOS'15, 2015. www.usenix.org/system/files/conference/hotos15/ hotos15-paper-mcsherry.pdf.

T. Mikolov, K. Chen, G. Corrado, and J. Dean. Efficient estimation of word representations in vector space. In Proceedings of ICLR Workshops Track, 2013. arxiv.org/abs/1301.3781.

V. Mnih, N. Heess, A. Graves, and K. Kavukcuoglu. Recurrent models of visual attention. In Proceedings of NIPS, pages 2204-2212, 2014. papers.nips.cc/paper/5542-recurrent-models-of-visual-attention.pdf.

V. Mnih, K. Kavukcuoglu, D. Silver, A. A. Rusu, J. Veness, M. G. Bellemare, A. Graves, M. Riedmiller, A. K. Fidjeland, G. Ostrovski, S. Petersen, C. Beattie, A. Sadik, I. Antonoglou, H. King, D. Kumaran, D. Wierstra, S. Legg, and D. Hassabis. Human-level control through deep reinforcement learning. Nature, 518(7540):529-533, 02 2015. dx.doi.org/10.1038/nature14236.

P. Moritz, R. Nishihara, I. Stoica, and M. I. Jordan. SparkNet: Training deep networks in Spark. In Proceedings of ICLR, 2016. arxiv.org/abs/1511.06051.

D. G. Murray, F. McSherry, M. Isard, R. Isaacs, P. Barham, and M. Abadi. Incremental, iterative data processing with timely dataflow. Commun. ACM, 59(10):75-83, Sept. 2016. dl.acm.org/citation.cfm?id=2983551.

A. Nair, P. Srinivasan, S. Blackwell, C. Alcicek, R. Fearon, A. De Maria, V. Panneershelvam, M. Suleyman, C. Beattie, S. Petersen, et al. Massively parallel methods for deep reinforcement learning. arXiv preprint, 1507.04296, 2015. arxiv.org/abs/1507.04296.

Nervana Systems. Neon deep learning framework, 2016. github.com/NervanaSystems/neon.

NVIDIA Corporation. NCCL: Optimized primitives for collective multi-GPU communication, 2016. github.com/NVIDIA/nccl.

R. Pascanu, T. Mikolov, and Y. Bengio. On the difficulty of training recurrent neural networks. In Proceedings of ICML, pages 1310-1318, 2013. jmlr.org/proceedings/papers/v28/pascanu13.pdf.

B. Recht, C. Re, S. Wright, and F. Niu. Hogwild: A lock-free approach to parallelizing stochastic gradient descent. In Proceedings of NIPS, pages 693-701, 2011. papers.nips.cc/paper/4390-hogwild-a-lock-free-approach-to-parallelizing-stochastic-gradient-descent.pdf.

C. J. Rossbach, Y. Yu, J. Currey, J.-P. Martin, and D. Fetterly. Dandelion: a compiler and runtime for heterogeneous systems. In Proceedings of SOSP, pages 49-68, 2013. sigops.org/sosp/sosp13/papers/p49-rossbach.pdf.

D. E. Rumelhart, G. E. Hinton, and R. J. Williams. Learning representations by back-propagating errors. In Cognitive modeling, volume 5, pages 213-220. MIT Press, 1988. www.cs.toronto.edu/~hinton/absps/naturebp.pdf.

O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla, M. Bernstein, A. C. Berg, and L. Fei-Fei. ImageNet Large Scale Visual Recognition Challenge. International Journal of Computer Vision, 115(3):211-252, 2015. arxiv.org/abs/1409.0575.

A. Smola and S. Narayanamurthy. An architecture for parallel topic models. Proc. VLDB Endow., 3(1-2):703-710, Sept. 2010. vldb.org/pvldb/vldb2010/papers/R63.pdf.

I. Sutskever, J. Martens, G. E. Dahl, and G. E. Hinton. On the importance of initialization and momentum in deep learning. In Proceedings of ICML, pages 1139-1147, 2013. jmlr.org/proceedings/papers/v28/sutskever13.pdf.

I. Sutskever, O. Vinyals, and Q. V. Le. Sequence to sequence learning with neural networks. In Proceedings of NIPS, pages 3104- 3112, 2014. papers.nips.cc/paper/5346-sequence-to-sequence-learning-with-neural.pdf.

C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich. Going deeper with convolutions. In Proceedings of CVPR, pages 1-9, 2015. arxiv.org/abs/1409.4842.

C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna. Rethinking the Inception architecture for computer vision. arXiv preprint, 1512.00567, 2015. arxiv.org/abs/1512.00567.

C. tao Chu, S. K. Kim, Y. an Lin, Y. Yu, G. Bradski, K. Olukotun, and A. Y. Ng. Map-reduce for machine learning on multicore. In Proceedings of NIPS, pages 281-288, 2007. papers.nips.cc/paper/3150-map-reduce-for-machine-learning-on-multicore.pdf.

A. Verma, L. Pedrosa, M. Korupolu, D. Oppenheimer, E. Tune, and J. Wilkes. Large-scale cluster management at Google with Borg. In Proceedings of EuroSys, 2015. research.google.com/pubs/archive/43438.pdf.

O. Vinyals, L. Kaiser, T. Koo, S. Petrov, I. Sutskever, and G. Hinton. Grammar as a foreign language. arXiv preprint, 2014. arxiv.org/abs/1412.7449.

Y. Wu, M. Schuster, Z. Chen, Q. V. Le, M. Norouzi, W. Macherey, M. Krikun, Y. Cao, Q. Gao, K. Macherey, J. Klingner, A. Shah, M. Johnson, X. Liu, L. Kaiser, S. Gouws, Y. Kato, T. Kudo, H. Kazawa, K. Stevens, G. Kurian, N. Patil, W. Wang, C. Young, J. Smith, J. Riesa, A. Rudnick, O. Vinyals, G. Corrado, M. Hughes, and J. Dean. Google's Neural Machine Translation system: Bridging the gap between human and machine translation. arXiv preprint, 1609.08144, 2016. arxiv.org/abs/1609.08144.

Y. Yu, M. Isard, D. Fetterly, M. Budiu, U. Erlingsson, P. K. Gunda, and J. Currey. DryadLINQ: A system for general-purpose distributed data-parallel computing using a high-level language. In Proceedings of OSDI, pages 1-14, 2008. www.usenix.org/legacy/event/osdi08/tech/full_papers/yu_y/yu_y.pdf.

M. Zaharia, M. Chowdhury, T. Das, A. Dave, J. Ma, M. McCauley, M. J. Franklin, S. Shenker, and I. Stoica. Resilient distributed datasets: A fault-tolerant abstraction for in-memory cluster computing. In Proceedings of NSDI, pages 15-28, 2012. https://www.usenix.org/system/files/conference/nsdi12/nsdi12-final138.pdf.

M. D. Zeiler, M. Ranzato, R. Monga, M. Mao, K. Yang, Q. Le, P. Nguyen, A. Senior, V. Vanhoucke, J. Dean, and G. E. Hinton. On rectified linear units for speech processing. In Proceedings of ICASSP, pages 3517-3521, 2013. research.google.com/pubs/archive/40811.pdf.