Chief AI Architect at UST
Data-Centric AI is the point of view around making it easier for practitioners to understand, program and iterate on datasets, instead of spending time on models. While contemporary AI and ML practices has been focused on models, the real-world experience of those who put models into production is that the data often matters more. The goal of this presentation is to explain the tenets of data centric AI such as Data Augmentation, Data tagging, Data Programming & Weak Supervision, Self-Supervision, and MLOps with a Data-centric approach to make it accessible by anyone who wants to understand and contribute to this area. Furthermore, we will explore how MLOps makes data centric AI an efficient and systematic process, and what role virtualized data fabric can play in Data-centric AI.