Scholar
Leman Akoglu
Google Scholar ID: 4ITkr_kAAAAJ
Associate Professor, Carnegie Mellon University
Citations & Impact
Citations
10,906
H-index
41
i10-index
95
Publications
20
Co-authors
44
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Multiple papers accepted at top venues including NeurIPS 2025, KDD 2025, JMLR, TMLR, AAAI/ACM AIES, SIGSPATIAL 2025, SIAM SDM 2025, AutoML 2024, IEEE BigData 2023, etc.
- Research topics include foundation models for zero-shot tabular outlier detection, discrete diffusion, synthetic mixed-prior pretraining for tabular FMs, human mobility anomaly detection, image anomaly detection with self-supervision, GPS trajectory anomaly detection, self-supervised time series OD, and algorithmic bias in OD models
- Invited keynote speaker at International Conference on Automated Machine Learning (AutoML) 2024
- Invited speaker at GraphEx Symposium, Fields Conference on Complex Networks in Banking and Finance (2024)
- Frequent invited speaker at international workshops and schools (2022–2023), including PFIA, Toronto Metropolitan University, and ISI Kolkata Winter School
- Work on detecting upcoding-driven waste in Medicare claims accepted to Journal of Policy Analysis and Management (2025)
Background
- Dean's Associate Professor (tenured) at Heinz College of Information Systems and Public Policy, Carnegie Mellon University
- Holds courtesy appointments in the Machine Learning Department (MLD) and Computer Science Department (CSD), School of Computer Science
- Directs the Data Analytics Techniques Algorithms (DATA) Lab at Heinz College
- Research interests broadly span data mining, graph mining, machine learning, and knowledge discovery
- Specific focus on anomaly detection ('anOmaLiEs')—identifying and characterizing outliers in large-scale, time-varying, multi-modal data using scalable computational methods
