George Kesidis

George Kesidis is a Professor in Computer Science and Engineering and Electrical Engineering at Penn State University. His research spans deep learning security, virtual reality optimization, and cloud computing.

• College of Engineering (Penn State University)
• Research Focus: Backdoor Attacks, DNN Robustness, Edge Caching
• Active in NSF and U.S. Navy-funded projects (2022-2026)

His work addresses backdoor data poisoning, test-time evasion attacks, and DNN overfitting mitigation. He develops techniques like activation clipping, perturbation analysis, and statistical defense models. Recent projects include edge caching systems for VR and security-driven AI frameworks.

Key article trends reveal expertise in adversarial deep learning, immersive media delivery, and cloud resource optimization. Current grants focus on multi-user VR, GPU scheduling, and serverless-cloud hybrid architectures.

He collaborates extensively with researchers like David J. Miller and Xinyu Li, particularly on cloud-based adversarial defense mechanisms and VR streaming benchmarks.