Haris Vikalo

Bio

I am a Professor in the Department of Electrical and Computer Engineering at The University of Texas at Austin. My research focuses on machine learning and signal processing methods for inference, learning, and decision-making from complex data, especially in settings where data is distributed, resources are limited, models must be adapted or compressed, or operating conditions change over time.

Current work in my group spans machine learning at the edge, federated and continual learning, and methods that use the geometry of representations and activations to understand, adapt, and compress large models. Applications include distributed learning systems, wireless communications, and computational genomics.

Research

• Machine learning at the edge. Algorithms and systems for learning and inference under latency, communication, computation, and memory constraints.
• Federated and continual learning. Distributed learning methods for heterogeneous, privacy-sensitive, and evolving data across clients, devices, and institutions.
• Wireless systems. Machine learning and signal processing for wireless communications, sensing, spectrum monitoring, and channel estimation.
• Computational genomics. Learning and inference methods for sequencing data, haplotype assembly, genomic mixture reconstruction, and biological response prediction.

More on selected research projects.

Recent Publications and Preprints (Full list: link.)

Links