Gleb Novikov

I am a machine learning researcher at the Lucerne University of Applied Sciences and Arts (HSLU). I received my PhD in Computer Science from ETH Zürich, where I was advised by David Steurer.

Theoretical Research

My theoretical research lies at the intersection of machine learning, algorithms, and statistics, with work on robust machine learning, statistical estimation, differential privacy, and computational aspects of learning. My research has been published at NeurIPS, ICML, COLT, FOCS, SODA, and ALT. See my publications for the full list.

I have served as a reviewer for NeurIPS, ICLR, ICML, and COLT, a subreviewer for STOC, and a Program Committee member for ALT and AAAI.

I currently have two manuscripts under submission and two ongoing research projects.

Applied Research

My applied research focuses on modern machine-learning methods for real-world problems, including large language models and document understanding, time-series forecasting, and interpretable sequential decision-making. I work on model development, empirical evaluation, and translating research ideas into practical ML systems. See my applied ML projects for more details.

Beyond my technical research, I work directly with industry stakeholders, translating real-world requirements into research problems, coordinating across technical and business teams, and communicating results through management reports, presentations, and client documentation.

Hobbies

I like ballroom dancing and enjoy dancing with my wife, Elena. We have competed in and won multiple dance competitions in Switzerland, and we are currently dancing in Class C (Latin).