Federated Learning with Position-Aware Neurons
This addresses the challenge of efficient model aggregation in Federated Learning for distributed systems, offering a simple, algorithm-agnostic solution that enhances performance without centralizing data.
The paper tackles the problem of parameter misalignment in Federated Learning due to neural network permutation invariance and non-i.i.d. data by proposing Position-Aware Neurons (PANs), which incorporate position encodings to pre-align parameters across clients, enabling effective coordinate-based averaging and improving existing FL algorithms.
Federated Learning (FL) fuses collaborative models from local nodes without centralizing users' data. The permutation invariance property of neural networks and the non-i.i.d. data across clients make the locally updated parameters imprecisely aligned, disabling the coordinate-based parameter averaging. Traditional neurons do not explicitly consider position information. Hence, we propose Position-Aware Neurons (PANs) as an alternative, fusing position-related values (i.e., position encodings) into neuron outputs. PANs couple themselves to their positions and minimize the possibility of dislocation, even updating on heterogeneous data. We turn on/off PANs to disable/enable the permutation invariance property of neural networks. PANs are tightly coupled with positions when applied to FL, making parameters across clients pre-aligned and facilitating coordinate-based parameter averaging. PANs are algorithm-agnostic and could universally improve existing FL algorithms. Furthermore, "FL with PANs" is simple to implement and computationally friendly.