MemComputing is a new physics-based approach to computation that employs memory (time non-locality) to both process and store information on the same physical location [1]. After a brief introduction to this computing paradigm and the physics principles behind it, I will discuss how it can be utilized in the field of Machine Learning, by showing efficient supervised and unsupervised training of neural networks, demonstrating its advantages over traditional methods and its usefulness in several science and engineering applications.
[1] M. Di Ventra, MemComputing: Fundamentals and Applications (Oxford University Press, 2022).
Massimiliano Di Ventra obtained his undergraduate degree in Physics summa cum laude from the University of Trieste (Italy) in 1991 and did his PhD studies at the Swiss Federal Institute of Technology in Lausanne in 1993-1997. Since 2004, he is a professor of Physics at the University of California, San Diego. Di Ventra has worked on a variety of topics in condensed matter physics such as the theory of quantum transport in nanoscale and atomic systems, non-equilibrium statistical mechanics, polymer dynamics/DNA sequencing in nanopores, and memory effects in nanostructures. Recently, he has introduced the MemComputing paradigm of computation. He has been invited to deliver more than 350 talks worldwide on these topics including 20 plenary/keynote presentations. He has published more than 300 papers in refereed journals, 5 textbooks, and has 12 granted patents (5 US, 7 foreign). He is a fellow of the American Association for the Advancement of Science, the American Physical Society, the Institute of Physics, the IEEE, a member of the National Academy of Inventors and of Academia Europaea. In 2018 he was named Highly Cited Researcher by Clarivate Analytics; he is the recipient of the 2024 Humboldt Research Award, the 2020 Feynman Prize for theory in Nanotechnology, and is a 2022-2023 IEEE Nanotechnology Council Distinguished Lecturer. He is the co-founder of MemComputing, Inc. (http://memcpu.com/).
MemComputing is a new physics-based approach to computation that employs memory (time non-locality) to both process and store information on the same physical location [1]. After a brief introduction to this computing paradigm and the physics principles behind it, I will discuss how it can be utilized in the field of Machine Learning, by showing efficient supervised and unsupervised training of neural networks, demonstrating its advantages over traditional methods and its usefulness in several science and engineering applications.
[1] M. Di Ventra, MemComputing: Fundamentals and Applications (Oxford University Press, 2022).
Massimiliano Di Ventra obtained his undergraduate degree in Physics summa cum laude from the University of Trieste (Italy) in 1991 and did his PhD studies at the Swiss Federal Institute of Technology in Lausanne in 1993-1997. Since 2004, he is a professor of Physics at the University of California, San Diego. Di Ventra has worked on a variety of topics in condensed matter physics such as the theory of quantum transport in nanoscale and atomic systems, non-equilibrium statistical mechanics, polymer dynamics/DNA sequencing in nanopores, and memory effects in nanostructures. Recently, he has introduced the MemComputing paradigm of computation. He has been invited to deliver more than 350 talks worldwide on these topics including 20 plenary/keynote presentations. He has published more than 300 papers in refereed journals, 5 textbooks, and has 12 granted patents (5 US, 7 foreign). He is a fellow of the American Association for the Advancement of Science, the American Physical Society, the Institute of Physics, the IEEE, a member of the National Academy of Inventors and of Academia Europaea. In 2018 he was named Highly Cited Researcher by Clarivate Analytics; he is the recipient of the 2024 Humboldt Research Award, the 2020 Feynman Prize for theory in Nanotechnology, and is a 2022-2023 IEEE Nanotechnology Council Distinguished Lecturer. He is the co-founder of MemComputing, Inc. (http://memcpu.com/).
