Weightless Neural Networks (WNN) are ideal for Federated Learning due to their robustness and computational efficiency. These scenarios require models with a small memory footprint and the ability to aggregate knowledge from multiple models. In this work, we demonstrate the effectiveness of using Bloom filter variations to implement DRASiW models—an adaptation of WNN that records both the presence and frequency of patterns—with minimized memory usage. Across various datasets, DRASiW models show competitive performance compared to models like Random Forest, -Nearest Neighbors, Multi-layer Perceptron, and Support Vector Machines, with an acceptable space trade-off. Furthermore, our findings indicate that Bloom filter variations, such as Count Min Sketch, can reduce the memory footprint of DRASiW models by up to 27% while maintaining performance and enabling distributed and federated learning strategies.

This work was partially supported by Fundação para a Ciência e a Tecnologia, I.P. (FCT) [ISTAR Projects: UIDB/04466/2020 and UIDP/04466/2020] and DSAIPA/AI/0122/2020 Aim Health. The authors would also like to thank CNPq (404087/2021-3 and 315399/2023-6)

Bloom filters,DRASiW,Knowledge aggregation,Weightless neural network

• Ciências da Computação e da Informação - Ciências Naturais

Esta publicação é um output do(s) seguinte(s) projeto(s):

• Aplicações Móveis Baseadas em Inteligência Artificial para Resposta de Saúde Pública
• Alternative Shaper
• Análise de Dados de Imagem e Vídeo
• O Impacto das Tecnologias Blockchains e de registo descentralizado