Federated Optimization:Distributed Optimization Beyond the Datacenter
Jakub Konečný, McMahan, Brendan, Daniel Ramage · arXiv (Cornell University) · 2015
We introduce a new and increasingly relevant setting for distributed optimization in machine learning, where the data defining the optimization are distributed (unevenly) over an extremely large number of \nodes, but the goal remains to train a high-quality centralized model. We refer to this setting as Federated Optimization. In this setting, communication efficiency is of utmost importance. A motivating example for federated optimization arises when we keep the training data locally on users' mobile devices rather than logging it to a data center for training. Instead, the mobile devices are used as nodes performing computation on their local data in order to update a global model. We suppose that we have an extremely large number of devices in our network, each of which has only a tiny fraction of data available totally; in particular, we expect the number of data points available locally to be much smaller than the number of devices. Additionally, since different users generate data with different patterns, we assume that no device has a representative sample of the overall distribution. We show that existing algorithms are not suitable for this setting, and propose a new algorithm which shows encouraging experimental results. This work also sets a path for future research needed in the context of federated optimization.
Cited by 544 works. doi:10.48550/arxiv.1511.03575 Open-access PDF
Authors
References
- Accelerating Stochastic Gradient Descent using Predictive Variance Reduction
- Communication Efficient Distributed Optimization using an Approximate Newton-type Method
- Communication-efficient algorithms for statistical optimization
- Distributed Coordinate Descent Method for Learning with Big Data
- Distributed Learning, Communication Complexity and Privacy
- Semi-Stochastic Gradient Descent Methods
- Trading Computation for Communication: Distributed Stochastic Dual Coordinate Ascent
- Communication Complexity of Distributed Convex Learning and Optimization
- Communication-Efficient Distributed Dual Coordinate Ascent
- Information-theoretic lower bounds for distributed statistical estimation with communication constraints