With the fast development of machine learning (especially deep learning) and cloud computing, it has become a trend to train machine learning models in a distributed manner on a cluster of machines. This workshop at AAAI 2017 (to be held on Feb 5, 2017) is designed to discuss important topics regarding distributed machine learning, from system, algorithm, and theory perspectives. We hope to provide the community with deep insights and to substantially push the frontier of distributed machine learning.

Invited talks

  • Alex Smola (Amazon): Why your machine learning algorithm is slow
  • Joseph E. Gonzalez (UC Berkeley): RISE to the Next Challenges of AI Systems
  • Xiangrui Meng (Databricks): Implementing large-scale matrix factorization on Apache Spark
  • Christopher Ré (Stanford University): Tuning large-scale systems: surprising system-algorithm interactions (to be presented by  Dr. Ioannis Mitliagka on behalf of Chris).

Abstracts of and slides of these talks could be found here.

Accepted papers

  • Ayan Das, Raghuveer Chanda, Smriti Agrawal, Sourangshu Bhattacharya, Distributed Weighted Parameter Averaging for SVM Training on Big Data
  • Soham De, Thomas Goldstein, CentralVR: A Framework for Variance-Reduced Distributed Optimization
  • Travis Dick, Mu Li, Venkata Krishna Pillutla, Colin White, Maria-Florina Balcan, Alex Smola, Data Driven Resource Allocation for Distributed Learning
  • Xi He, Dheevatsa Mudigere, Mikhail Smelyanskiy, Martin Takác, Distributed Hessian-Free Optimization for Deep Neural Network
  • Chenxin Ma, Martin Takác, Distributed Inexact Damped Newton Method: Data Partitioning and Work-Balancing
  • Praveen Rao, Anas Katib, Kobus Barnard, Charles Kamhoua, Kevin Kwiat, Laurent L. Njilla, Scalable Score Computation for Learning Multinomial Bayesian Networks over Distributed Data
  • Jun Song, David A. Moore, Parallel Chromatic MCMC with Spatial Partitioning
  • Gavin Taylor, Zheng Xu, Thomas Goldstein, Scalable Classifiers with ADMM and Transpose Reduction
  • Romain Warlop, Alessandro Lazaric, Jérémie Mary, Parallel Higher Order Alternating Least Square for Tensor Recommender System

A tentative schedule has been available here.

Call for papers

Important dates

  • Paper submission deadline: Oct 21, 2016 (submission closed)
  • Notification: Nov 18, 2016
  • Camera ready due: Dec 8, 2016
  • Workshop date: Feb 5, 2017