Data Science Tea
Speaker: Natesh Ganesh
The shift in the focus of the computing industry towards learning applications, and the inevitable end to Moore's law (`death by CMOS scaling and heat') have empowered the search for unconventional substrates, to build computing systems that can "learn like the human brain." While we have made tremendous progress in realizing intelligence in narrow tasks using large datasets and compute power, artificial general intelligence (AGI) that can rival the human brain in terms of energy efficiency and performance across domains continues to be the holy grail of computing. To progress from narrow to general AI, it is important to have a solid theoretical foundation for computing and intelligence, and what they both entail. These are important questions to address, to help us identify the optimal devices, architectures and design techniques that will allow us to efficiently build the intelligent systems of the future. This talk will review the fundamental ideas and assumptions that have allowed us to achieve computing over the years. There are crucial distinctions between our intelligence, and that achieved through current computational approaches, which begs the foundational question - is computing the optimal path to intelligence? Building off these ideas, I will use recent results from non-equilibrium thermodynamics to propose an alternate theoretical framework of thermodynamic intelligence, that treats intelligence as a physical process, describes it in terms of homeostasis, entropy flow and energy dissipation, and does not espouse to computing. And discuss the path moving forward - introduction of a new engineering paradigm to realize thermodynamic intelligence called thermodynamic computing, change to existing design philosophies, and novel technologies that might serve as good testbeds.
Natesh Ganesh is a PhD student in the Electrical and Computer Engineering Dept. at Umass, Amherst working under Prof. Neal Anderson. He is interested in physical limits to computing, brain inspired hardware, non-equilibrium thermodynamics, emergence of intelligence in self-organized systems, theories of consciousness, and philosophy of mind. He was awarded the best paper award at IEEE ICRC'17 for the paper - "A Thermodynamic Treatment of Intelligent Systems". He proposed the new physical framework of thermodynamic intelligence, and helped define the engineering paradigm of thermodynamic computing, and organize the first workshop on the research are (https://cra.org/ccc/events/thermodynamic-computing/).