CollaFuse: Navigating Limited Resources and Privacy in Collaborative Generative AI

arXiv:2402.19105v1 Announce Type: new
Abstract: In the landscape of generative artificial intelligence, diffusion-based models present challenges for socio-technical systems in data requirements and privacy. Traditional approaches like federated learning distribute the learning process but strain individual clients, especially with constrained resources (e.g., edge devices). In response to these challenges, we introduce CollaFuse, a novel framework inspired by split learning. Tailored for efficient and collaborative use of denoising diffusion probabilistic models, CollaFuse enables shared server training and inference, alleviating client computational …

abstract artificial artificial intelligence arxiv challenges collaborative cs.ai cs.lg data devices diffusion edge edge devices federated learning generative generative artificial intelligence intelligence landscape privacy process requirements resources systems technical type