DISCO: A Browser-Based Privacy-Preserving Framework for Distributed Collaborative Learning

Julien Tuấn Tú Vignoud, Martin Jaggi, Mary-Anne Hartley, Tahseen Rabbani, Valérian Rousset

Keywords: federated learning, decentralized learning, collaborative learning, machine learning, deep learning, browser, smartphone, privacy, confidentiality

TL;DR: DISCO enables training deep learning models collaboratively directly in the browser, without sharing raw data.

Abstract: Data is often impractical to share for a range of well considered reasons, such as concerns over privacy, intellectual property, and legal constraints. This not only fragments the statistical power of predictive models, but creates an accessibility bias, where accuracy becomes inequitably distributed to those who have the resources to overcome these concerns. We present **DISCO**: an open-source **Dis**tributed **Co**llaborative learning platform accessible to non-technical users, offering a means to collaboratively build machine learning models without sharing any original data or requiring any programming knowledge. DISCO's web application trains models locally directly in the browser, making our tool cross-platform out-of-the-box, including smartphones. The modular design of DISCO offers choices between federated and decentralized paradigms, various levels of privacy guarantees and several approaches to weight aggregation strategies that allow for model personalization and bias resilience in the collaborative training.

Submission Number: 3