Dr Narayan Srinivasa of Arch Systems, USA

Modern AI systems compress reality into hidden mathematical spaces called latent representations—invisible dimensions that hold the key to how machines understand and create. This talk traces the evolution from early approaches, such as VAEs, to today's revolutionary diffusion models, which power Stable Diffusion and beyond.

We'll decode the mathematical elegance behind diffusion processes: how learning to reverse noise systematically unlocks unprecedented creative capabilities. Through latent diffusion architectures, we'll explore how these systems achieve remarkable efficiency by operating in compressed spaces rather than raw pixels, enabling real-time human-AI collaboration.

Beyond technical foundations, we'll examine how diffusion-powered latent representation is reshaping creative industries, scientific discovery, and everyday applications. As these invisible spaces become increasingly controllable, they're transforming how humans and AI collaborate, making the invisible visible through intelligent partnership.

Biography:

Narayan Srinivasa received his Ph. D. from the University of Florida in Gainesville in 1994. He was a Beckman Post-Doctoral Fellow in the Human-Computer Intelligent Interaction group at the Beckman Institute, University of Illinois at Urbana-Champaign, from 1994 to 1997. Between 1998 and 2015, he was with HRL Laboratories in Malibu, CA, where he became Principal Scientist and the Director for Neural and Emergent Systems. At HRL, he worked on a wide range of AI projects, including visual perception and computer vision, signal processing and sensor fusion, brain-inspired computing, and robotics. He joined Intel Labs in 2016 as Chief Scientist to lead the development of neuromorphic technology and played a key role in the development of the Loihi neuromorphic chip. Until October of 2025, he was Senior Principal AI Engineer and Director for Machine Intelligence Research Programs at Intel Labs. Since November 2025, he has been Chief AI Scientist at Arch Systems, where he is responsible for accelerating AI research in NLP, Computer Vision, and Data processing and analytics. He holds 126 issued U.S. patents, with several patents pending, and has authored over 104 articles in journals, magazines, and conferences. He was a TEDx speaker in 2012 on intelligent machines. He is a Fellow of the IEEE and a distinguished IEEE Lecturer for the Computational Intelligence Society.

Address:Arch Systems, , California, United States