SAN JOSE, California - UTRC's Soumalya Sarkar was one of three invited speakers who shared insights regarding intelligent machines, artificial intelligence and deep learning as part of the Industrial Internet of Things session at the GPU Technology Conference held earlier this week in Silicon Valley.
A Senior Research Scientist, Decision Support and Machine Intelligence, in the Systems Department, Sarkar delivered a talk, titled, "Deep Representation and Reinforcement Learning for Anomaly Detection and Control in Multi-modal Aerospace Applications."
Research relating to this topic was conducted jointly by Sarkar and other members of the Systems team: Kishore Reddy, Amit Surana, Vivek Venugopalan, Mark Gurvich and Michael Giering.
Sarkar's 50-minute presentation focused on how deep auto-encoder (DAE) and deep reinforcement learning (DRL) can be formulated to address multimodal anomaly detection and additive manufacturing control problems within the aerospace industry.
"The GPU Technology Conference serves as a global forum to connect with technology experts in industry, academia and the government," said Sarkar. "It also offers valuable training and a place to showcase the most vital work in the computing industry today -- including artificial intelligence and deep learning, industrial Internet of Things (IIOT) and intelligent machines, healthcare, virtual reality, accelerated analytics and multi-physics simulation and self-driving cars. I was very pleased to be invited to share UTRC’s game-changing developments in this space."
Contact: Laura Stevens, (860) 610-1653