Alexander Terenin

Alexander Terenin is an Assistant Research Professor at Cornell University, specializing in machine learning and artificial intelligence. His work focuses on decision-making under uncertainty, Bayesian optimization, and Gaussian processes, particularly in non-Euclidean spaces. He has contributed to geometric learning, scalable Gaussian process methods, and applications in robotics, plasma science, and legal AI.

His research integrates theoretical foundations with practical algorithms, emphasizing principles like the Gittins Index for optimal decision-making. Notable projects include the GeometricKernels software package for manifold learning and the Cambridge Law Corpus for legal AI. His work bridges statistics, geometry, and computer science to address challenges in autonomous systems, energy optimization, and data-driven decision-making.

Recent Talks and Contributions:

• An Adversarial Analysis of Thompson Sampling (INFORMS Applied Probability Society 2025)
• Cost-aware Bayesian Optimization (NeurIPS 2024)
• Stochastic Poisson Surface Reconstruction (ICML 2025)

Key Research Themes:

• Bayesian Optimization for multi-objective problems (e.g., plasma-driven energy systems)
• Geometric Gaussian Processes for robotics and 3D modeling
• Statistical guarantees for Gaussian processes on manifolds

Grants and Collaborations: His work involves interdisciplinary projects with institutions like Carnegie Mellon University, ETH Zürich, and the University of Cambridge, reflecting a global network in AI and statistical learning.