Dr. Sumit Jha Investigates New Pathways for Fabrication of In-Memory Computing Systems

Dr. Sumit Jha Investigates New Pathways for Fabrication of In-Memory Computing Systems

KFSCIS Research Professor Sumit Jha, has launched a new, collaborative project designed to bring together expertise in formal methods, machine learning, computer-aided design, and fabrication of in-memory computing (IMC) systems. The project’s goal is to create formal methods that can synthesize neural networks in the memory of the computer and also prove their correctness.

“We proposed a very novel idea for the research,” Jha stated. “By leveraging recent innovations in machine learning and formal methods, this project synthesizes crossbars for neural nets using decision diagrams, neural nets, and reinforcement learning. It verifies bi-directional digital IMC circuits before demonstrating such in-memory computing systems through fabrication.”

The project will also include the verification of neural networks accelerated using analog in-memory computing (IMC) and the synthesis of hybrid analog-digital IMC for neural networks using formal methods and machine learning, while demonstrating these innovations using infield fabrication of the IMC systems.

These innovations will be critical in allowing the training of neural networks with reduced power consumption and are particularly important with the larger adoption of AI and the need to train more powerful neural networks.

Additional project contributions will include, enhancing the reliability of neural networks on in memory circuits, increasing diversity and computer engineering and computer science, and fostering interdisciplinary collaboration across formal methods, machine learning, and hardware design.

The project, titled, “Collaborative Research: FMitF: Track I: Synthesis and Verification of In-Memory Computing Systems using Formal Methods”, is a collaboration between the University of Central Florida, SUNY Polytechnic Institute, and FIU. The three-year award totals $750,000, with $250,000 going to each institution, and is funded by the National Science Foundation. Additional information can be found at https://www.nsf.gov/awardsearch/showAward?AWD_ID=2404036 .

// Live helper chat widget begin // Live helper chat widget end