Smart Reply: Automated Response Suggestion for Email

In this paper we propose and investigate a novel end-to-end method for automatically generating short email responses, called Smart Reply. It generates semantically diverse suggestions that can be used as complete email responses with just one tap on mobile. The system is currently used in Inbox by Gmail and is responsible for assisting with 10% of all mobile responses. It is designed to work at very high throughput and process hundreds of millions of messages daily. The system exploits state-of-the-art, large-scale deep learning. We describe the architecture of the system as well as the challenges that we faced while building it, like response diversity and scalability. We also introduce a new method for semantic clustering of user-generated content that requires only a modest amount of explicitly labeled data.

[1]  References , 1971 .

[2]  Slava M. Katz,et al.  Estimation of probabilities from sparse data for the language model component of a speech recognizer , 1987, IEEE Trans. Acoust. Speech Signal Process..

[3]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[4]  Zoubin Ghahramani,et al.  Combining active learning and semi-supervised learning using Gaussian fields and harmonic functions , 2003, ICML 2003.

[5]  Alexander Zien,et al.  Label Propagation and Quadratic Criterion , 2006 .

[6]  Xiao Li,et al.  Understanding the Semantic Structure of Noun Phrase Queries , 2010, ACL.

[7]  Yoram Singer,et al.  Adaptive Subgradient Methods for Online Learning and Stochastic Optimization , 2011, J. Mach. Learn. Res..

[8]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[9]  Alan Ritter,et al.  Data-Driven Response Generation in Social Media , 2011, EMNLP.

[10]  Michael Gamon,et al.  Predicting Responses to Microblog Posts , 2012, NAACL.

[11]  Bo Pang,et al.  Revisiting the Predictability of Language: Response Completion in Social Media , 2012, EMNLP-CoNLL.

[12]  Meichun Hsu,et al.  Identifying Intention Posts in Discussion Forums , 2013, NAACL.

[13]  Jon M. Kleinberg,et al.  Characterizing and curating conversation threads: expansion, focus, volume, re-entry , 2013, WSDM.

[14]  Andrew W. Senior,et al.  Long short-term memory recurrent neural network architectures for large scale acoustic modeling , 2014, INTERSPEECH.

[15]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[16]  Quoc V. Le,et al.  Sequence to Sequence Learning with Neural Networks , 2014, NIPS.

[17]  Rishiraj Saha Roy,et al.  Discovering and understanding word level user intent in Web search queries , 2015, J. Web Semant..

[18]  Jianfeng Gao,et al.  A Neural Network Approach to Context-Sensitive Generation of Conversational Responses , 2015, NAACL.

[19]  Joelle Pineau,et al.  Hierarchical Neural Network Generative Models for Movie Dialogues , 2015, ArXiv.

[20]  Hang Li,et al.  Neural Responding Machine for Short-Text Conversation , 2015, ACL.

[21]  Samy Bengio,et al.  Show and tell: A neural image caption generator , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Quoc V. Le,et al.  Listen, Attend and Spell , 2015, ArXiv.

[23]  Sujith Ravi,et al.  Large Scale Distributed Semi-Supervised Learning Using Streaming Approximation , 2015, AISTATS.

[24]  Joelle Pineau,et al.  Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models , 2015, AAAI.

[25]  Yoav Goldberg,et al.  A Primer on Neural Network Models for Natural Language Processing , 2015, J. Artif. Intell. Res..

[26]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[27]  Marc-Allen Cartright,et al.  Hierarchical Label Propagation and Discovery for Machine Generated Email , 2016, WSDM.