Feb 20, 2026. Congratulations to Farzad Ordubadi and Michael Keyser for their paper being accepted in IEEE ISCAS’2026, Shanghai, China. The paper is entitled: “A Discrete-Time Analog Compute Macro for Edge ML with 200 TOPS/W in CMOS 28nm”. This article explores new ways to implement analog Multiply and Accumulate (MAC) circuits for implementing neural networks.
Relevant work: Keyser, Michael, Farzad Ordubadi, and Armin Tajalli. “Effect of Parameter Drift on the Accuracy of Analog CNN Accelerators.” 2024 IEEE 67th International Midwest Symposium on Circuits and Systems (MWSCAS). IEEE, 2024.