Michael Sherman

Rutgers, The State University of New Jersey, University of Chicago

Papers

Total Citations

H-Index

About

Michael Sherman’s research sits at the intersection of optical communications, machine learning, and distributed systems, with a focus on real-world sensing and education. His most cited work, “Efficient Classification of Polarization Events Based on Field Measurements” (2020, 8 citations), pioneers rare-event classification of polarization transients using data augmentation and robot-generated fiber-disturbance data—a critical contribution for securing optical networks against physical tampering. By systematically comparing machine learning methods, he established practical guidelines for balancing accuracy with training sample efficiency. His more recent “AutoLearn: Learning in the Edge to Cloud Continuum” (2023, 4 citations) addresses the growing need for hands-on experimentation in cloud computing and autonomous systems, leveraging NSF-supported testbeds to democratize access. This work reflects his broader commitment to bridging theory and practice, particularly in education. Sherman’s impact is amplified by his ability to translate complex field measurements into deployable classification frameworks, making him a key figure in both optical sensing and cyber-physical systems education. His research continues to shape how we detect, classify, and learn from rare events in real-world environments.