Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Leader Selection and Follower Association for UE-centric Distributed Learning in Future Wireless Networks

Saeedeh Parsaeefard, Sabine Roessel, Anousheh Gholami Ghavamabad, Robert Zaus, Bernhard Raaf

TL;DR

Two new indices are introduced that transform the highly complex leader selection and follower association problem into a better tractable formulation and allow to keep the internal and external state information of this problem inside of each device without compromising users’ privacy.

Abstract

User equipment (UE) devices with high compute performance acting on data with dynamic and stochastic nature to train Artificial Intelligence/Machine Learning (AI/ML) models call for real-time, agile distributed machine learning (DL) algorithms. Consequently, we focus on UE-centric DL algorithms where UEs initiate requests to adapt AI/ML models for better performance, e.g., locally refined AI/ML models among a set of headsets or smartphones. This new setup requires selecting a set of UEs as aggregators (here called leaders) and another set as followers, where all UEs update their models based on their local data, and followers share theirs with leaders for aggregation. From a networking perspective, the first question is how to select leaders and associate followers efficiently. This results in a high dimensional mixed integer programming problem and involves internal UE state information and state information among UEs, called external state information in this paper. To address this challenge, we introduce two new indices: a Leader Internal Index (LII), which is a function of the internal states of each device, demonstrating the willingness to be a leader such as battery life and AI/hardware accelerators, and a Leader eXternal Index (LXI), which is a function of external state information among UEs, such as trust, channel condition, and any aspect relevant for associating a follower with a leader. These two indices transform the highly complex leader selection and follower association problem into a better tractable formulation. More importantly, LIIs and LXIs allow to keep the internal and external state information of this problem inside of each device without compromising users' privacy.

Leader Selection and Follower Association for UE-centric Distributed Learning in Future Wireless Networks

TL;DR

Two new indices are introduced that transform the highly complex leader selection and follower association problem into a better tractable formulation and allow to keep the internal and external state information of this problem inside of each device without compromising users’ privacy.

Abstract

User equipment (UE) devices with high compute performance acting on data with dynamic and stochastic nature to train Artificial Intelligence/Machine Learning (AI/ML) models call for real-time, agile distributed machine learning (DL) algorithms. Consequently, we focus on UE-centric DL algorithms where UEs initiate requests to adapt AI/ML models for better performance, e.g., locally refined AI/ML models among a set of headsets or smartphones. This new setup requires selecting a set of UEs as aggregators (here called leaders) and another set as followers, where all UEs update their models based on their local data, and followers share theirs with leaders for aggregation. From a networking perspective, the first question is how to select leaders and associate followers efficiently. This results in a high dimensional mixed integer programming problem and involves internal UE state information and state information among UEs, called external state information in this paper. To address this challenge, we introduce two new indices: a Leader Internal Index (LII), which is a function of the internal states of each device, demonstrating the willingness to be a leader such as battery life and AI/hardware accelerators, and a Leader eXternal Index (LXI), which is a function of external state information among UEs, such as trust, channel condition, and any aspect relevant for associating a follower with a leader. These two indices transform the highly complex leader selection and follower association problem into a better tractable formulation. More importantly, LIIs and LXIs allow to keep the internal and external state information of this problem inside of each device without compromising users' privacy.
Paper Structure (13 sections, 5 equations, 5 figures, 2 tables, 1 algorithm)

This paper contains 13 sections, 5 equations, 5 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Motivation and Novelties
  3. Literature Review
  4. Paper Organization
  5. System Model
  6. Joint Leader Selection and Follower Association Formulation
  7. Distributed scenario
  8. Architecture Aspects and Signaling for Future Wireless Networks
  9. Simulation Setup and Results
  10. Model for LII and LXI
  11. Benchmarking the Distributed Algorithm
  12. Cluster Configurations
  13. Conclusion

Figures (5)

  • Figure 1: System model of $N=3$ UEs in the coverage of a base station (BS) equipped with edge server, denoted by index $0$
  • Figure 2: Process of proposing edge server to cover marginal scenarios in distributed algorithm.
  • Figure 3: The network procedure and signaling for UE-centric DL algorithm in 5G and 6G
  • Figure 4: Cluster configurations from left to right: (a) Exhaustive Search, (b) Phase 1 of the Distributed Algorithm for an appropriate value of $\rho$, (c) Phase 1 of the Distributed Algorithm for a too small value of $\rho$, (d) Phase 2 for the too small value of $\rho$. All figures for an identical set of LIIs and LXIs.
  • Figure 5: Distributed algorithm’s (a) mean utility value and (b) mean (resulting) leader set size across thresholds $\rho \in \{0,1\cdots, 9\}$ for $N=10$ compared to optimal solution’s mean, (c) leader set size probability distribution (average over all $\rho$) of distributed algorithm and of optimal solution.