Abstract: Nature uses collective analog and probabilistic computation with noisy and unreliable devices to achieve incredible efficiency in energy, time, and space, e.g, in the brain or in cells. I will show how inspiration from nature and precise analog circuit engineering can architect novel ultra-energy-efficient and supercomputing architectures. Examples include: 1) Analog neural prosthetics that are so energy efficient that they could be powered by glucose in the body; 2) Analog synthetic biological operational amplifiers in living microbial cells that could release smart drugs with precise feedback control;  3) Analog cytomorphic chips for arbitrary drug discovery in medicine that simulate complex and noisy biological networks with orders-of-magnitude acceleration; 4) Classical analog circuits on VLSI chips that emulate arbitrary quantum Hamiltonians with exponential tensor-product superpositions but using only linear scaling. The latter chips are important for A.I. and other applications that need to operate at room temperature. One can even simultaneously combine inspiration from biology and physics, e.g., to create a Quantum Cochlea for spectral analysis of light or radio signals rather than sound. We need to compute more efficiently like Nature does, blurring boundaries between energy and information, physics and computation. Analog and Probabilistic Computers provide a means for doing so. 

Bio: Rahul Sarpeshkar obtained Bachelor’s degrees in Electrical Engineering and Physics at M.I.T and a PhD in Computation and Neural Systems at the California Institute of Technology. Right after his PhD, he was a Member of the Technical Staff in Bell Labs Physics Department in its division of Biological Computation. His book, Ultra Low Power Bioelectronics: Fundamentals, Biomedical Applications, and Bio-inspired Systems was written at M.I.T., where he was a tenured professor, and a professor in Electrical Engineering and Computer Science. He is currently the Thomas E. Kurtz Professor at Dartmouth, and a Professor of Engineering, Professor of Physics, Professor of Microbiology and Immunology, and Professor of Molecular and Systems Biology. He holds 44 awarded patents and is a Fellow of the National Academy of Inventors and a Fellow of the IEEE. His talk will describe work in four fields that he has pioneered, all of which involve analog, biological, or quantum computing.