Work

Deep Randomized Distributed Function Computation (DeepRDFC): Neural Distributed Channel Simulation

Year: 2025

Type: conference-paper

Abstract: The randomized distributed function computation (RDFC) framework, which unifies many cutting-edge distributed computation and learning applications, is considered. An autoencoder (AE) architecture is proposed to minimize the total variation distance between the probability distribution simulated by the AE outputs and an unknown target distribution, using only data samples. We illustrate significantly high RDFC performance with communication load gains from our AEs compared to data compression methods. Our designs establish deep learning-based RDFC methods and aim to facilitate the use of RDFC methods, especially when the amount of common randomness is limited and strong function computation guarantees are required.

Source: IEEE International Symposium on Information Theory (ISIT)

DOI: 10.1109/isit63088.2025.11195680

Authors: Didrik Bergström, Onur Günlü

Institution: Linköping University

Topic: Distributed Sensor Networks and Detection Algorithms

Subfield: Computer Networks and Communications

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: Stiftelsen för Strategisk Forskning, National Science Council, Linköpings Universitet, National Supercomputer Centre, Linköpings Universitet

OpenAlex ID: W4415367662

FWCI: 3.864

Cites: 22

Cited by: 3

Related to: