Federated Gaussian Process: Convergence, Automatic Personalization and Multi-fidelity Modeling

11/28/2021
by   Xubo Yue, et al.
9

In this paper, we propose : a Federated Gaussian process (𝒢𝒫) regression framework that uses an averaging strategy for model aggregation and stochastic gradient descent for local client computations. Notably, the resulting global model excels in personalization as jointly learns a global 𝒢𝒫 prior across all clients. The predictive posterior then is obtained by exploiting this prior and conditioning on local data which encodes personalized features from a specific client. Theoretically, we show that converges to a critical point of the full log-likelihood function, subject to statistical error. Through extensive case studies we show that excels in a wide range of applications and is a promising approach for privacy-preserving multi-fidelity data modeling.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 1

page 2

page 3

page 4

09/16/2021

Federated Submodel Averaging

We study practical data characteristics underlying federated learning, w...
07/15/2020

Tackling the Objective Inconsistency Problem in Heterogeneous Federated Optimization

In federated optimization, heterogeneity in the clients' local datasets ...
03/02/2021

PFA: Privacy-preserving Federated Adaptation for Effective Model Personalization

Federated learning (FL) has become a prevalent distributed machine learn...
10/11/2020

Federated Learning via Posterior Averaging: A New Perspective and Practical Algorithms

Federated learning is typically approached as an optimization problem, w...
07/31/2021

A graphical Gaussian process model for multi-fidelity emulation of expensive computer codes

We present a novel Graphical Multi-fidelity Gaussian Process (GMGP) mode...
06/20/2021

Is Shapley Value fair? Improving Client Selection for Mavericks in Federated Learning

Shapley Value is commonly adopted to measure and incentivize client part...
11/29/2019

Richer priors for infinitely wide multi-layer perceptrons

It is well-known that the distribution over functions induced through a ...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.