Volkan Cevher

About

Volkan Cevher is a leading researcher in machine learning and optimization, with key contributions spanning high-dimensional Bayesian optimization, inverse reinforcement learning, and large-scale algorithmic efficiency. His work on "High-Dimensional Bayesian Optimization via Additive Models with Overlapping Groups" (2018, 34 citations) addresses a critical challenge in sequential black-box function optimization—scaling to high dimensions—by introducing additive structures that enable tractable and effective exploration. This has broad implications for parameter tuning, robotics, and environmental monitoring. In "Interaction-limited Inverse Reinforcement Learning" (2020, 3 citations), Cevher tackles a practical yet underexplored problem: accelerating learning when teacher-learner interaction is constrained, such as in scenarios where a helpful teacher is unavailable or cannot provide continuous feedback. His research consistently bridges theoretical rigor with real-world applicability, advancing the frontiers of optimization and learning under resource limitations. Cevher’s work is widely recognized for its impact on algorithmic design, and he continues to shape the field through innovative frameworks that push the boundaries of data-efficient and interaction-aware machine learning.

Research Focus

Key Achievements

Top Papers

1. 1
   High-Dimensional Bayesian Optimization via Additive Models with Overlapping Groups

   34 citations · 2018
2. 2
   Interaction-limited Inverse Reinforcement Learning

   3 citations · 2020

Key Collaborators

Contact & Links