Privacy-preserving federated learning AdaBoost via knowledge distillation
Knowledge-Based Systems, Vol. 339, Article 116962, 2026
Federated Learning (FL) provides a decentralized machine learning paradigm allowing multiple clients to collaboratively train models without sharing raw data. However, data heterogeneity across clients and privacy risks during model exchanges remain critical challenges. In this paper, we propose a novel privacy-preserving adaptation of AdaBoost for Federated Learning that leverages knowledge distillation to effectively mitigate non-IID data skew while preventing sensitive information leakage during aggregation, enhancing both model accuracy and robust privacy guarantees.
BAM-SLDK: Biologically inspired Attention Mechanism with Spiking Learnable Delayed Kernel synapses
IOP Neuromorphic Computing and Engineering, Vol. 5, No. 1, 2025
Spiking Neural Networks (SNNs) are emerging as an alternative neural network model due to their biological plausibility, energy efficiency, and built-in ability to learn from temporal dynamics. However, in order to effectively process data with rich spatial and temporal dependencies, the usual static projections (feedforward and recurrent) among layers of spiking neurons fail to represent all the information needed. Inspired by how synaptic delays affect the learning process in biological neurons, in this paper, we propose a biologically inspired attention mechanism based on spiking convolutions with learnable delayed kernel synapses. The proposed model increases temporal learning ability, attending simultaneously to spatial and temporal dynamics with few parameters required.