CRANE
Zero-shot anomaly detection for recognizing defects when labeled abnormal examples are limited.
Trustworthy artificial intelligence
Building AI that remains reliable beyond the benchmark.
I am an Associate Professor at New Uzbekistan University and a Visiting Researcher at OIST, Japan. My work focuses on anomaly detection, out-of-distribution detection, AI robustness, safety, and learning under unknown conditions.
Associate ProfessorNew Uzbekistan University
Visiting ResearcherOIST, Japan
Principal InvestigatorJSPS-funded AI research
Area ChairICLR 2027
Selected work
Three research directions connecting machine-learning foundations to dependable systems in the real world.
FrameShield
Robust video anomaly detection designed to remain dependable under adversarial visual inputs.
Universal Novelty Detection
Detecting previously unseen data across datasets and tasks through adaptive contrastive learning.
AI consulting for industry
Independent technical advice, robustness audits, computer vision and anomaly-detection strategy, due diligence, workshops, and focused proofs of concept.
Latest updates
The complete archive remains available on the News page.
Serving as an Area Chair for ICLR 2027.
CRANE accepted at BMVC 2026.
Joined New Uzbekistan University as an Associate Professor.