Λέσχη Φίλων Στατιστικής - GrStats forum
AUEB Stats Seminars 2/12/2022: Federated Bayesian Computation via Piecewise Deterministic Markov Processes (Delft Institute of Applied Mathematics, NL) Forumgrstats

Join the forum, it's quick and easy

Λέσχη Φίλων Στατιστικής - GrStats forum
AUEB Stats Seminars 2/12/2022: Federated Bayesian Computation via Piecewise Deterministic Markov Processes (Delft Institute of Applied Mathematics, NL) Forumgrstats
Λέσχη Φίλων Στατιστικής - GrStats forum
Would you like to react to this message? Create an account in a few clicks or log in to continue.
Για προβλήματα εγγραφής και άλλες πληροφορίες επικοινωνήστε με : grstats.forum@gmail.com ή grstats@stat-athens.aueb.gr

Go down
avatar
GRStats2
Posts : 4
Join date : 2022-11-19

AUEB Stats Seminars 2/12/2022: Federated Bayesian Computation via Piecewise Deterministic Markov Processes (Delft Institute of Applied Mathematics, NL) Empty AUEB Stats Seminars 2/12/2022: Federated Bayesian Computation via Piecewise Deterministic Markov Processes (Delft Institute of Applied Mathematics, NL)

Tue 29 Nov 2022 - 20:01
ΚΥΚΛΟΣ ΣΕΜΙΝΑΡΙΩΝ ΣΤΑΤΙΣΤΙΚΗΣ ΔΕΚΕΜΒΡΙΟΣ 2022

Joris Bierkens
Delft Institute of Applied Mathematics, NL

Federated Bayesian Computation via Piecewise Deterministic Markov Processes

FRIDAY 2/12/2022
15:00

Teams Link
https://www.facebook.com/events/2968869493418122?ref=newsfeed

ABSTRACT

When performing Bayesian computations in practice, one is often faced with the challenge that the constituent model components and/or the data are only available in a distributed fashion, e.g. due to privacy concerns or sheer volume. While various methods have been proposed for performing posterior inference in such federated settings, these either make very strong assumptions on the data and/or model or otherwise introduce significant bias when the local posteriors are combined to form an approximation of the target posterior. By leveraging recently developed methods for Markov Chain Monte Carlo (MCMC) based on Piecewise Deterministic Markov Processes (PDMPs), we develop a computation- and communication- efficient family of posterior inference algorithms (Fed-PDMC) which provides asymptotically exact approximations of the full posterior over a large class of Bayesian models, allowing heterogenous model and data contributions from each client. We show that communication between clients and the server preserves the privacy of the individual data sources by establishing differential privacy guarantees. We quantify the performance of Fed-PDMC over a class of illustrative analytical case-studies and demonstrate its efficacy on a number of synthetic examples along with realistic Bayesian computation benchmarks.

Joint work with Andrew Duncan (Imperial College London)
https://arxiv.org/abs/2210.13816
Back to top
Permissions in this forum:
You cannot reply to topics in this forum