Federated Learning: How to train AI models in the age of sensitive data
Standard machine learning approaches require centralizing the training data on one machine or in a datacenter. Federated Learning enables training models on distributed and private datasets without the need to centralize them. Privacy and security are key considerations for data set owners participating in Federated Learning optimizations.
Walter Riviera is AI Technical Specialist EMEA Lead at Intel.
Walter joined Intel in 2017 as an AI TSS (Technical Solution Specialist) covering EMEA and he’s now playing an active role on most of the AI project engagements within the Data Centers business in Europe. He is responsible for increasing Technical and business awareness regarding the Intel AI Offer, enabling and provide technical support to end user customers, ISVs, OEMs, Partners in implementing HPC and/or Clouds solutions for AI based on Intel’s products and technologies. Before joining Intel Walter has collected research experiences working on adopting ML techniques to enhance images retrieval algorithms for robotic applications, conducting sensitive data analysis in a start-up environment and developing software for Text To Speech applications.
