Proceedings of the 4th International Workshop on Distributed Machine Learning

摘要

Following up the prior three successful versions of DistributedML, it is our great honour and pleasure to welcome you again, this time physically in the 4rd edition of the Distributed Machine Learning Workshop (DistributedML '23). The workshop is co-located with the 19th International Conference on emerging Networking EXperiments and Technologies (CoNEXT '23) and held in Paris, France, on the 8th of December 2023. Distributed ML is a rapidly evolving, interdisciplinary field bringing together techniques from Distributed Systems, Networks and Machine Learning. With Deep Learning at the forefront, we are seeing an explosion of AI-driven technologies, from immersive VR experiences and smart digital assistants to advanced robotics and autonomous vehicles. These applications not only challenge the limits of local device capabilities but also many times necessitate a shift towards distributed models of computation to enhance performance and efficiency, while respecting privacy and sustainability. At the cornerstone of innovation, foundational models further push the boundaries of today's computational infrastructure. Therefore, scaling up to support the new training workloads and efficiently deploying Large Language or Vision Models become key research areas.