Neuromorphic Computing – An area that enforces the principles of neurology to computing systems to mimic the function and structure of the brain – if it needs to scale it on the scale to compete effectively with current computing methods Is. One in Publish reviews on 22 January in Journal Nature23 researchers, including two of California University San Diego, presented a detailed roadmap of what needs to be reached to reach that goal. The article provides a new and practical perspective towards getting closer to the cognitive ability of the human brain with comparable form factor and electric consumption.
“We do not guess that the scale will have a shape-fit-all solution for neuromorphic systems, but a range of neuromorphic hardware solutions with various characteristics based on the needs of the application,” the author writes.
Applications for neuromorphic computing include scientific computing, Artificial Intelligence, Augmented and Virtual Reality, Wearbals, Smart Farming, Smart City and more. Neuromorphic chips have the ability to beat traditional computers in energy and space efficiency, as well as performance. It can offer adequate benefits in various domains including AI, health care and robotics. As AI has electricity consumption Doubled by 2026Neuromorphic computing emerges as a promising solution.
“Neuromorphic computing is especially relevant today, when we are looking at the volatile scaling and resource-buzzing AI systems of electrical,” said UC San Diego Shoo Chion-Jee-Jean Lap Department of Bioinizing and one of the one Girt Covenbergs. Paper’s coates.
Neuromorphic computing is in a significant moment, Robert F at the University of Texas San Antonio. McDermot -rich chairman, Dhirshe Kudithipudi and the concerned writer of the paper said. “We are now at a point where there is a tremendous opportunity to create new architecture and open structures that can be deployed in commercial applications,” he said. âI strongly believe that promoting tight cooperation between industry and academics is the key to shaping the future of the region. This cooperation is reflected in our team of co-author. ,
Last year, Cauwenberghs and Kudithipudi received a grant of $ 4 million to launch from National Science Foundation Thor: Neuromorphic CommonsA first-type research network that provides access to neuromorphic computing hardware and equipment opening in support of interdisciplinary and collaborative research.
In 2022, a neuromorphic chip designed by a team led by Cauwenberghs showed that these chips could be highly dynamic and versatile, without compromising accuracy and efficiency. Nerram chip Directly calculates in memory and AI can run a wide variety of applications-all the general-obvious AI computing for computing computing platforms on a fraction of energy. “Our nature review article provides a perspective on further expansion of neuromorphic AI system in silicon and large-scale emerging chip technologies Presented to, “Covenbergs said.
To achieve the scale in neuromorphic computing, the author proposes several key features that must be adapted, including sparsity, a defined characteristic of the human brain. Most of them develop several nerve connections (density) before sorting them selectively. This strategy optimizes spatial efficiency while maintaining knowledge on high loyalty. If successfully imitated, this feature can enable neuromorphic systems that are quite high energy-skilled and compact.
“Extension scalability and better efficiency are derived from mass equality and hierarchical structure in nerve representation, mixing dense local synaptic connectivity within the neurosinaptic core mixed with dense local synaptic connectivity within the neurosinaptic core and after modeling of the brain after modeling of the brain With, modeling the white substance of the brain, through the high, provided through the high.
âThis publication shows tremendous ability towards the use of neuromorphic computing on a scale for real -life applications. At the San Diego Supercomputer Center, we bring new computing architecture to the national user community, and this collaborative work paves the way to bring a neuromorphic resources to the national user community, â Amitawa Majumdar, Director of Data-Cappected Scientific Computing Division at SDSC at the UC San Diego campus, and one of the co-workers of paper.
In addition, the author also calls for strong cooperation within academics and between academics and industry, as well as for the development of a broad array of user -friendly programming languages ââto reduce the barrier of entry into the region. They believe it will promote increased cooperation, especially in subjects and industries.
Dhiresh Kudithipudi and Tej Pandit, University of Texas, San Antonio
Catherine Shuman, University of Tennessi, Knocville
Craig M. Wineyard, James B.Corey Merkel, Rochastor Institute of Technology
Prince Kubendran, University of Pittsburgh
Garrick Orchard and Rayad Benosman, Intel Labs
Christian Mayor, Technish Universe Dresseden
Joe Haze, US Naval Research Laboratory,
Cliff Young, Google Deepmind
Chiara bartolozy
Amitawa Majumdar and Gart Covenberg, California San Diego University
Melika Pavand, Neuroinformatics Institute, University of Zurich and Ath Zurich
Sonia Bakle, National Institute of Standards and Technology
Shruti Kulkarni, Oak Ridge National Laboratory
Hector A. Gonzalez, Spinchloud System GMBH, DressDen, Germany
Chetan Singh Thakur, Indian Institute of Science, Bengaluru
Anand Subramani, Royal Holove, London University, Aghham
Steve Forms, University of Manchester
Diego
