About Konstantin Burlachenko Konstantin Burlachenko Ph.D. Student, Computer Science Federated learning Optimization for Machine Learning High Performance Computing software development Konstantin's research focuses on large-scale mathematical optimization systems for machine learning and artificial intelligence, especially those requiring high-performance computing approaches. Events Presented Events May 11 - May 17, 2025 Error Feedback for Communication-Efficient First and Second-Order Distributed Optimization: Theory and Practical Implementation Konstantin Burlachenko, Ph.D. Student, Computer Science May 12, 12:00 - 13:00 B9 L2 R2325 Federated learning software development This seminar will discuss advancements in Federated Learning, including theoretical improvements to the Error Feedback method (EF21) for communication-efficient distributed training and the development of significantly more practical and efficient implementations of the Federated Newton Learn (FedNL) algorithm. May 4 - May 10, 2025 Optimization Methods and Software for Federated Learning Konstantin Burlachenko, Ph.D. Student, Computer Science May 8, 19:00 - 21:00 B5 L5 R5209 This dissertation identifies five key challenges in Federated Learning (FL), including data and device heterogeneity, communication issues, privacy concerns, and software implementations. More broadly, our work serves as a guide for researchers navigating the complexities of translating theoretical methods into efficient real-world implementations, while also offering insights into the reverse process of adapting practical implementation aspects back into theoretical algorithm design.
Error Feedback for Communication-Efficient First and Second-Order Distributed Optimization: Theory and Practical Implementation Konstantin Burlachenko, Ph.D. Student, Computer Science May 12, 12:00 - 13:00 B9 L2 R2325 Federated learning software development This seminar will discuss advancements in Federated Learning, including theoretical improvements to the Error Feedback method (EF21) for communication-efficient distributed training and the development of significantly more practical and efficient implementations of the Federated Newton Learn (FedNL) algorithm.
Optimization Methods and Software for Federated Learning Konstantin Burlachenko, Ph.D. Student, Computer Science May 8, 19:00 - 21:00 B5 L5 R5209 This dissertation identifies five key challenges in Federated Learning (FL), including data and device heterogeneity, communication issues, privacy concerns, and software implementations. More broadly, our work serves as a guide for researchers navigating the complexities of translating theoretical methods into efficient real-world implementations, while also offering insights into the reverse process of adapting practical implementation aspects back into theoretical algorithm design.
Engage LinkedIn GitHub ORCID ShareClipboard Related Sites Center of Excellence for Generative AI (GenAI) Optimization and Machine Learning (OML) Computer Science (CS) Related Content Events 2 Related Links Homepage Curriculum vitae Google Scholar