WASP project - Deployment and automated tuning of ML functions in distributed RAN environments
Title: |
WASP project - Deployment and automated tuning of ML functions in distributed RAN environments |
DNr: |
Berzelius-2022-10 |
Project Type: |
LiU Berzelius |
Principal Investigator: |
Martin Isaksson <martin.isaksson@gmail.com> |
Affiliation: |
Kungliga Tekniska högskolan |
Duration: |
2022-01-20 – 2022-08-01 |
Classification: |
10201 |
Homepage: |
https://wasp-sweden.org/ |
Keywords: |
|
Abstract
This project will do research on distributed and decentralized machine learning methods, for example Federated Learning (FL), gossip learning, transfer learning and reinforcement learning - specifically the aspects of such functions that apply to a distributed cloud/virtual 5G RAN environment.
FL is a promising framework for distributed learning when data is private
and sensitive. However, the state-of-the-art solutions in this framework are
not optimal when data is heterogeneous and non-IID. FL, especially device-based FL is also resource intensive due to the very large amount of clients needed to investigate these methods.