SCAFFOLD: Stochastic Controlled Averaging for Federated Learning
Jul 12, 2020
Speakers
Praneeth Karimireddy
Speaker · 0 followers
Satyen Kale
Speaker · 0 followers
Mehryar Mohri
Speaker · 4 followers
Sashank Jakkam Reddi
Speaker · 0 followers
Sebastian Stich
Speaker · 0 followers
Ananda Theertha Suresh
Speaker · 0 followers
About
Federated learning is a key scenario in modern large-scale machine learning where the data remains distributed over a large number of clients and the task is to learn a centralized model without transmitting the client data. The standard optimization algorithm used in this setting is Federated Averaging (FedAvg) due to its low communication cost. We obtain a tight characterization of the convergence of FedAvg and prove that heterogeneity (non-iid-ness) in the client's data results in a `drift' in the local updates resulting in poor performance. As a solution, we propose a new algorithm (SCAFFOLD) which uses control variates (variance reduction) to correct for the `client drift'. We prove that SCAFFOLD requires singificantly fewer communication rounds and is not affected by data heterogeneity or client sampling. Further, we show that (for quadratics) SCAFFOLD can take advantage of similarity in the client's data yielding even faster convergence. The latter is the first result to quanitfy the usefulness of local-steps in distributed optimization.Federated learning is a key scenario in modern large-scale machine learning where the data remains distributed over a large number of clients and the task is to learn a centralized model without transmitting the client data. The standard optimization algorithm used in this setting is Federated Averaging (FedAvg) due to its low communication cost. We obtain a tight characterization of the convergence of FedAvg and prove that heterogeneity (non-iid-ness) in the client's data results in a `drift' i…
Categories
AI & Data Science
Category · 10.8k presentations
About ICML 2020
The International Conference on Machine Learning (ICML) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence known as machine learning. ICML is globally renowned for presenting and publishing cutting-edge research on all aspects of machine learning used in closely related areas like artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, and robotics. ICML is one of the fastest growing artificial intelligence conferences in the world. Participants at ICML span a wide range of backgrounds, from academic and industrial researchers, to entrepreneurs and engineers, to graduate students and postdocs.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
New Oracle-Efficient Algorithms for Private Synthetic Data Release
Watch later
Invited Talk 5 Q&A
Watch later
Retrospective Publish ´Til You Perish
Watch later
ConQUR: Mitigating Delusional Bias in Deep Q-Learning
Watch later
DiJia Su, …
Double-Loop Unadjusted Langevin Algorithm
Watch later
Paul Rolland, …
