arXiv subjects

Yingxin Liu

Publications and source records attributed to Yingxin Liu.

3 recordsLinked to original sources

arXiv2025

Adaptive Client Selection via Q-Learning-based Whittle Index in Wireless Federated Learning

We consider the client selection problem in wireless Federated Learning (FL), with the objective of reducing the total required time to achieve a certain level of learning accuracy. Since the server cannot observe the clients' dynamic states that can change their computation and communication efficiency, we formulate client selection as a restless multi-armed bandit problem. We propose a scalable and efficient approach called the Whittle Index Learning in Federated Q-learning (WILF-Q), which uses Q-learning to adaptively learn and update an approximated Whittle index associated with each client, and then selects the clients with the highest indices. Compared to existing approaches, WILF-Q does not require explicit knowledge of client state transitions or data distributions, making it well-suited for deployment in practical FL settings. Experiment results demonstrate that WILF-Q significantly outperforms existing baseline policies in terms of learning efficiency, providing a robust and efficient approach to client selection in wireless FL.

cs.LG↗

arXiv2023

A Ring Topology-based Communication-Efficient Scheme for D2D Wireless Federated Learning

Federated learning (FL) is an emerging technique aiming at improving communication efficiency in distributed networks, where many clients often request to transmit their calculated parameters to an FL server simultaneously. However, in wireless networks, the above mechanism may lead to prolonged transmission time due to unreliable wireless transmission and limited bandwidth. This paper proposes a communication scheme to minimize the uplink transmission time for FL in wireless networks. The proposed approach consists of two major elements, namely a modified Ring All-reduce (MRAR) architecture that integrates D2D wireless communications to facilitate the communication process in FL, and a modified Ant Colony Optimization algorithm to identify the optimal composition of the MRAR architecture. Numerical results show that our proposed approach is robust and can significantly reduce the transmission time compared to the conventional star topology. Notably, the reduction in uplink transmission time compared to baseline policies can be substantial in scenarios applicable to large-scale FL, where client devices are densely distributed.

math.OC↗

arXiv2022

Fisher information analysis on post-selection involved quantum precision measurements using optical coherent states

The weak-value-amplification (WVA) technique has been extensively considered and debated in the field of quantum precision measurement, largely owing to the reduced Fisher information caused by the low probability of successful post-selection. %% In this work we show that, rather than the Gaussian meter state as typically considered, using the optical coherent state as a meter, the WVA measurement can definitely outperform the conventional measurement not involving the strategy of post-selection. %% We also show that the post-selection procedure involved in the WVA scheme can make a mixture of coherent states work better than a pure coherent state with identical average photon numbers. This is in sharp contrast to the claim proved in the absence of post-selection. The post-selection strategy can also result in the precision of Heisenberg (or even "super-Heisenberg") scaling with the photon numbers, but without using any expensive quantum resources. %% The present work may stimulate further investigations for the potential of the post-selection strategy in quantum precision measurements.

quant-ph↗