Igor Fedorov

Papers

Total Citations

H-Index

About

Igor Fedorov is a researcher whose work lies at the intersection of machine learning, signal processing, and dictionary learning—a field focused on efficiently representing complex data through sparse, interpretable structures. His most cited paper, "Dictionaries in machine learning" (2023), provides a comprehensive survey of how learned representations can enhance model performance, bridging theoretical foundations with practical applications in compression, denoising, and feature extraction. Though early in his citation trajectory, Fedorov’s contributions are notable for their clarity and synthesis, offering a valuable resource for students and practitioners navigating the growing role of dictionary-based methods in modern AI. His work underscores the importance of interpretable, data-driven representations in advancing machine learning, particularly in domains where efficiency and transparency are critical. As his research continues to evolve, Fedorov is poised to make further impacts on how algorithms learn from and represent high-dimensional data.