The Database Lab at UC San Diego is one of the leading academic research groups in the field of
data management, spanning the major themes of theory, systems, languages, interfaces, and applications,
as well as intersections with other data-oriented fields, especially data science and AI.
Areas of particular strength include data analytics, semistructured and graph data,
data-centric AI, AI engineering, machine learning systems, causal inference, responsible data science,
query processing and optimization, and data exploration.
Application areas of particular interest have included healthcare, social media, and Internet of Things.
Our members span the departments of Computer Science and Engineering
and Halıcıoğlu Data Science Institute.
DB Lab faculty are also affiliated with other research groups, including the
HDSI Data Infrastructure and Systems Group,
HDSI AI and ML Group,
CSE AI Group,
and Center for Networked Systems.
Alin's research interests include data publishing and integration, specification and verification of DB-powered business processes and semistructured and XML data.
Arun's research interests are in data management and systems for ML/AI-based data analytics. His work focuses on designing abstractions, algorithms, and systems to make it easier and faster to analyze large and complex datasets using ML/AI.
Babak's research interests are in data management, causal inference, responsible data science, and data ethics. His work unifies techniques from DB theory, causal inference, and ML to lay the foundations for decision-making and policy evaluation from complex relational data, algorithmic fairness, explainability, and accountability.
Victor's research interests are in DB systems and theory. His current work focuses on verification of DB-driven systems, at the intersection of DB and computer-aided verification, and on automatic verification of interactive data-driven Web services and business processes. He is also interested in the theory of query languages and computational logic.
Yannis' research extends the capabilities of data platforms and query processors. He has published over 100 research articles with more than 14,000 citations.
Baharan's research interests lie in fair and explainable machine learning, debiasing the data, data cleaning for ML, and causal inference in relational data using different techniques like graph representation learning.
Jiongli's research interests lie in data cleaning and debiasing for machine learning applications.
Ilkay Altintas is the Director for the Center of Excellence in Workflows for Data Science at the San Diego Supercomputer Center (SDSC), UCSD.
Amarnath Gupta is a Research Scientist at the San Diego Supercomputer Center (SDSC) of the University of California San Diego.
June 2026
Arun promoted to Full Professor at UC San Diego. Congrats, Arun!
April 2026
Xiuwen, Arun, and Amarnath receive the Best Paper Award at IEEE ICDE 2026. Congrats!
February 2026
Victor retires from UC San Diego and moves to INRIA, France. We will miss you, Victor!
December 2025
Kyle defends his thesis and graduates with his PhD. Congrats and best wishes to him for her career!
June 2025
Xiuwen walks at the PhD commencement. Congrats and best wishes to her for her career!
SKALPEL: Schema Knowledge Adjustments for LLM Performance Enhancements in Large Databases
Kyle Luoma and Arun Kumar
2026 ; VLDB
MICRO: A Lightweight Middleware for Optimizing Cross-store Cross-model Graph-Relation Joins
Xiuwen Zheng, Arun Kumar, and Amarnath Gupta
2026 ; ICDE
Link to paper
SNAILS: Schema Naming Assessments for Improved LLM-based SQL Inference
Kyle Luoma and Arun Kumar
2025 ; SIGMOD
Link to paper
Causal DAG Summarization
Anna Zeng, Michael Cafarella, Batya Kenig, Markos Markakis, Brit Youngmann, and Babak Salimi
2025 ; VLDB
Link to paper
Stress-Testing ML Pipelines with Adversarial Data Corruption
Jiongli Zhu, Geyang Xu, Felipe Lorenzi, Boris Glavic, and Babak Salimi
2025 ; VLDB
Link to paper
Navigating Data Errors in Machine Learning Pipelines: Identify, Debug, and Learn
Bojan Karlaš, Babak Salimi, and Sebastian Schelter
2025 ; SIGMOD Tutorial
Link to paper
Zorro: Quantifying Uncertainty in Models & Predictions Arising from Dirty Data
Kaiyuan Hu, Jiongli Zhu, Boris Glavic, and Babak Salimi
2025 ; SIGMOD Demo
Link to paper
A Lightweight Method to Disrupt Memorized Sequences in LLM
Parjanya Prajakta Prashant, Kaustubh Ponkshe, and Babak Salimi
2025 ; NeurIPS
Link to paper
KAIROS: Scalable Model-Agnostic Data Valuation
Jiongli Zhu, Parjanya Prajakta Prashant, Alex Cloninger, and Babak Salimi
2025 ; NeurIPS
Link to paper
Scalable Out-of-distribution Robustness in the Presence of Unobserved Confounders
Parjanya Prashant, Seyedeh Baharan Khatami, Bruno Ribeiro, and Babak Salimi
2025 ; AISTATS
Link to paper
