معرفی
Vladimir Cherkassky serves as a Professor in the Department of Electrical and Computer Engineering at the University of Minnesota, with his office located at 6-111 Kenneth H. Keller Hall in Minneapolis. His research centers on predictive learning—specifically pattern recognition, statistical learning theory, and artificial neural networks—which intersects with data mining, statistical estimation, signal processing, and artificial intelligence.
Cherkassky's work emphasizes both theoretical foundations and practical applications, with fingerprint analysis revealing core expertise in Support Vector Machines (100%), Neural Networks (42%), Training Data (32%), and Model Selection (28%). His research bridges machine learning theory with real-world biomedical and engineering challenges, particularly in seizure prediction systems and materials science.
Recent publications (2020-2024) demonstrate a clear trend toward applying VC theory to deep learning's double descent phenomenon and developing clinical prediction tools using intracranial EEG data. These works consistently connect theoretical machine learning advances to applications in neuroscience and materials discovery.
Scientific Awards:
- No specific awards are documented in the provided materials
Grants and Projects: Cherkassky has secured four major research grants including NIH-funded projects Neurophysiologically Based Responsive Brain Tracking & Modulation (2015-2019) and Reliable Seizure Prediction Using Physiological Signals (2015-2020), an NSF project on Alternative Learning Methods (2008-2013), and a Minnesota Department of Transportation initiative for Soil Chemical Survey Data Analysis (2008-2010). His grant portfolio demonstrates sustained funding for translational machine learning research with clinical and engineering applications.
