Poster
Bayesian Nonparametric Federated Learning of Neural Networks
Mikhail Yurochkin · Mayank Agarwal · Soumya Ghosh · Kristjan Greenewald · Nghia Hoang · Yasaman Khazaeni
In federated learning problems, data is scattered across different servers and exchanging or pooling it is often impractical or prohibited. We develop a Bayesian nonparametric framework for federated learning with neural networks. Each data server is assumed to provide local neural network weights, which are modeled through our framework. We then develop an inference approach that allows us to synthesize a more expressive global network without additional supervision, data pooling and with as few as a single communication round. We then demonstrate the efficacy of our approach on federated learning problems simulated from two popular image classification datasets.
Author Information
Mikhail Yurochkin (IBM Research AI)
Mayank Agarwal (IBM Research)
Soumya Ghosh (IBM Research)
Kristjan Greenewald (IBM)
Nghia Hoang (MIT-IBM Watson AI Lab, IBM Research)
Yasaman Khazaeni (IBM Research)
Related Events (a corresponding poster, oral, or spotlight)
-
2019 Oral: Bayesian Nonparametric Federated Learning of Neural Networks »
Wed Jun 12th 03:15 -- 03:20 PM Room Hall A
More from the Same Authors
-
2019 Poster: Collective Model Fusion for Multiple Black-Box Experts »
Minh Hoang · Nghia Hoang · Bryan Kian Hsiang Low · Carleton Kingsford -
2019 Poster: Estimating Information Flow in Deep Neural Networks »
Ziv Goldfeld · Ewout van den Berg · Kristjan Greenewald · Igor Melnyk · Nam Nguyen · Brian Kingsbury · Yury Polyanskiy -
2019 Oral: Estimating Information Flow in Deep Neural Networks »
Ziv Goldfeld · Ewout van den Berg · Kristjan Greenewald · Igor Melnyk · Nam Nguyen · Brian Kingsbury · Yury Polyanskiy -
2019 Oral: Collective Model Fusion for Multiple Black-Box Experts »
Minh Hoang · Nghia Hoang · Bryan Kian Hsiang Low · Carleton Kingsford -
2018 Poster: Structured Variational Learning of Bayesian Neural Networks with Horseshoe Priors »
Soumya Ghosh · Jiayu Yao · Finale Doshi-Velez -
2018 Oral: Structured Variational Learning of Bayesian Neural Networks with Horseshoe Priors »
Soumya Ghosh · Jiayu Yao · Finale Doshi-Velez
