In this episode of the Practical AI podcast, Daniel Whitenack and Chris Benson interview Chong Shen Ng, a research engineer at Flower Labs, about federated learning and the Flower framework. Chong discusses his background in computational physics and how he transitioned to federated learning, highlighting its importance for handling sensitive and massive datasets. He explains federated learning as a method of training models on decentralized data sources without moving the data itself, emphasizing the user-friendliness and production capabilities of the Flower framework. The conversation covers the framework's architecture, deployment strategies, and its increasing relevance in training large language models, as well as the impact of generative AI on the roadmap for Flower.
