Abstract
Nowadays, Internet of things (IoT) devices and mobile phones are being used for collecting large volumes of data (e.g., photos, videos, location information) from various applications such as social media apps, medical equipment, and business platforms. Since the data is distributed and privacy-sensitive, the cloud-centric approaches for training a machine learning model may not be feasible, where the raw data is collected from many clients to train a model. Also, cloud-centric methods involve longer propagation delays and incur unacceptable latency for applications in which on-the-edge real-time decisions have to be made. Collaborative learning (CL) is a decentralized Machine Learning (ML) approach that trains machine learning models in a distributed way by guaranteeing that the training data remains on personal devices that preserve privacy and enable on-edge real-time decisions. In this talk, I am going to talk about collaborative learning and its applications in process industries. Amarlingam Madapu, ABB Bangalore
M. Amarlingam received the B.Tech. degree in electronics and communication engineering from Jawaharlal Nehru Technological University Hyderabad, India, in 2012, and the M.Tech. and Ph.D. degrees in electrical engineering from IIT Hyderabad in 2019. He worked as a Postdoctoral Researcher with the Department of Electrical and Communication Engineering, IISc, Bengaluru, India from 2019 to 2021. He is currently working as a scientist at INCRC, ABB Bangalore.