Accepted Papers


(Links to the camera-ready papers will be available online closer to the conference date.)


Beyond Data and Model Parallelism for Deep Neural Networks
Zhihao Jia (Stanford University), Matei Zaharia (Stanford University), Alex Aiken (Stanford University)

Restructuring Batch Normalization to Accelerate CNN Training
Wonkyung Jung (Seoul National University), Daejin Jung (Samsung), Byeongho Kim (Seoul National University), Sunjung Lee (Seoul National University), Wonjong Rhee (Seoul National University), Jung Ho Ahn (Seoul National University)

Optimizing DNN Computation With Relaxed Graph Substitutions
Zhihao Jia (Stanford University), James Thomas (Stanford University), Todd Warszawski (Stanford University), Mingyu Gao (Stanford University), Matei Zaharia (Stanford University), Alex Aiken (Stanford University)

Bandana: Using Non-Volatile Memory for Storing Deep Learning Models
Assaf Eisenman (Stanford University), Darryl Gardner (Facebook), Maxim Naumov (Facebook, Inc.), Misha Smelyanskiy (Facebook), Sergey Pupyrev (Facebook), Kim Hazelwood (Facebook), Uladzimir Pashkevich (Facebook), Asaf Cidon (Stanford University), Sachin Katti (Stanford University)

3LC: Lightweight and Effective Traffic Compression for Distributed Machine Learning
Hyeontaek Lim (Carnegie Mellon University), David G Andersen (Carnegie Mellon University), Michael Kaminsky (Intel Labs)

RLgraph: Flexible Computation Graphs for Deep Reinforcement Learning
Michael Schaarschmidt (University of Cambridge), Sven Mika (rlcore), Kai Fricke (Helmut Schmidt University), Eiko Yoneki (University of Cambridge)

AGGREGATHOR: Byzantine Machine Learning
Georgios Damaskinos (EPFL), El Mahdi El Mhamdi (EPFL), Rachid Guerraoui (EPFL), Arsany Guirguis (EPFL), Sébastien Rouault (EPFL)

FixyNN: Energy-Efficient Real-Time Mobile Computer Vision Hardware Acceleration via Transfer Learning
Paul Whatmough (ARM Research), Chuteng Zhou (Arm Research), Patrick Hansen (Arm Research), Shreyas Venkataramanaiah (Arizona State University), Jae-sun Seo (Arizona State University), Matthew Mattina (ARM Research)

Pytorch-BigGraph: A Large Scale Graph Embedding System
Adam Lerer (Facebook AI Research), Ledell Wu (Facebook AI Research), Jiajun Shen (Facebook AI Research), Alex Peysakhovich (Facebook), Timothee Lacroix (Facebook), Abhijit Bose (Facebook)

Priority-Based Parameter Propagation for Distributed DNN Training
Anand Jayarajan (University of British Columbia), Jinliang Wei (Carnegie Mellon University), Garth Gibson (Carnegie Mellon University), Alexandra Fedorova (University of British Columbia), Gennady Pekhimenko (University of Toronto)

Discrete Attacks and Submodular Optimization With Applications to Text Classification
Qi Lei (UT Austin), Lingfei Wu (IBM T. J. Watson Research Center), Pin-Yu Chen (IBM Research), Alex Dimakis (UT Austin), Inderjit Dhillon (University of Texas at Austin), Michael J Witbrock (IBM Research)

Serving Recurrent Neural Networks Efficiently on Spatial Architecture
Tian Zhao (Stanford University), Yaqi Zhang (Stanford University), Kunle Olukotun (Stanford University)

TensorFlow Eager: A Multi-Stage, Python-Embedded DSL for Machine Learning
Akshay Agrawal (Google), Alexandre Passos (Google), Allen Lavoie (Google), Igor Ganichev (Google), Akshay Modi (Google), Ashish Agarwal (Google), Asim Shankar (Google)

Ternary Hybrid Neural-Tree Networks for Highly Constrained IoT Applications
Dibakar Gope (Arm Inc.), Ganesh Dasika (Arm Research), Matthew Mattina (ARM Research)

CaTDet: Cascaded Tracked Detector for Efficient Object Detection From Video
Huizi Mao (stanford university), Taeyoung Kong (Stanford University), William J. Dally (Stanford University)

Adaptive Communication Strategies to Achieve the Best Error-Runtime Trade-Off in Local-Update SGD
Jianyu Wang (Carnegie Mellon University), Gauri Joshi (Carnegie Mellon University)

To Compress or Not to Compress: Understanding the Interactions Between Adversarial Attacks and Neural Network Compression
Ilia Shumailov (University of Cambridge), Yiren Zhao (University of Cambridge), Robert Mullins (University of Cambridge), Ross Anderson (University of Cambridge)

BlueConnect: Decomposing All-Reduce for Deep Learning on Heterogeneous Network Hierarchy
Minsik Cho (IBM Research), Ulrich Finkler (IBM Research), David Kung (IBM Research)

DropBack: Full DNN Training on a Pruned Weight Budget
Mieszko Lis (University of British Columbia), Maximilian Golub (UBC / Mercedes Benz Research & Dev.), Guy Lemieux (University of British Columbia)

Mini-Batch Serialization: CNN Training With Inter-Layer Data Reuse
Sangkug Lym (The University of Texas at Austin), Armand Behroozi (The University of Michigan), Wei Wen (Duke University), Ge Li (The University of Texas at Austin), Yongkee Kwon (University of Texas at Austin), Mattan Erez (UT Austin)

ParMAC: Distributed Optimisation of Nested Functions, With Application to Learning Binary Autoencoders
Miguel A Carreira-Perpinan (UC Merced), Mehdi Alizadeh (UC Merced)

YellowFin and the Art of Momentum Tuning
Jian Zhang (Stanford University), Ioannis Mitliagkas (University of Montreal)

TensorFlow.js: Machine Learning for the Web and Beyond
Daniel Smilkov (Google), Nikhil Thorat (Google), Yannick Assogba (Google), Charles Nicholson (Verily), Nick Kreeger (Google), Ping Yu (Google), Shanqing Cai (Google), Eric Nielsen (Google), David Soegel (Google), Stan Bileschi (Google), Michael Terry (Google), Ann Yuan (Google), Kangyi Zhang (Google), Sandeep Gupta (Google), Sarah Sirajuddin (Google), D Sculley (Google), Rajat Monga (Google), Greg Corrado (Google), Fernanda Viegas (Google), Martin M Wattenberg (Google)

Continuous Integration of Machine Learning Models: A Rigorous Yet Practical Treatment
Cedric Renggli (ETH Zurich), Bojan Karlaš (ETH Zürich), Bolin Ding ("Data Analytics and Intelligence Lab, Alibaba Group"), Feng Liu (Huawei Technologies), Kevin Schawinski (Modulos AG), Wentao Wu (Microsoft Research), Ce Zhang (ETH)

Data Validation for Machine Learning
Neoklis Polyzotis (Google), Martin Zinkevich (Google), Sudip Roy (Google), Eric Breck (Google), Steven Whang (KAIST)

Accurate and Efficient 2-Bit Quantized Neural Networks
Jungwook Choi (IBM Research), Swagath Venkataramani (IBM Research), Vijayalakshmi (Viji) Srinivasan (IBM TJ Watson), Kailash Gopalakrishnan (IBM Watson), Zhuo Wang (IBM Research), Pierce Chuang

Learning Kernels That Adapt to GPUs
Siyuan Ma (The Ohio State University), Mikhail Belkin (Ohio State University)

Towards Federated Learning at Scale: System Design
Wolfgang Grieskamp (Google), Brendan McMahan (Google), Vlivan Ivanov (Google)

AutoGraph: Imperative-style Coding with Graph-based Performance
Dan Moldovan (Google Inc.), James Decker (Purdue University), Fei Wang (Purdue University), Andrew Johnson (Google Inc.), Brian Lee (Google Inc.), Zack Nado (Google Inc.), D Sculley (Google), Tiark Rompf (Purdue University), Alexander B Wiltschko (Google Inc.)

Scaling Video Analytics on Constrained Edge Nodes
Christopher Canel (Carnegie Mellon University), Thomas Kim (Carnegie Mellon University), Giulio Zhou (Carnegie Mellon University), Conglong Li (Carnegie Mellon University), Hyeontaek Lim (Carnegie Mellon University), David G Andersen (Carnegie Mellon University), Michael Kaminsky (Intel Labs), Subramanya Dulloor (Intel Labs)

TicTac: Improving Distributed Deep Learning With Communication Scheduling
Sayed Hadi Hashemi (University of Illinois at Urbana-Champaign), Sangeetha Abdu Jyothi (University of Illinois at Urbana-Champaign), Roy Campbell (University of Illinois at Urbana-Champaign)

AdaScale: Towards Real-Time Video Object Detection Using Adaptive Scaling
Ting-Wu Chin (Carnegie Mellon University), Ruizhou Ding (Carnegie Mellon University), Diana Marculescu (Carnegie Mellon University)

Top