Active Projects

ADALab's research is organized under two top-level projects that focus on two different aspects of the ADA lifecycle:


Project Triptych
An end-to-end data system for sourcing data and features, as well as specifying, optimizing, and managing the ML model selection process.


Project Genisys
Deep learning-powered database perception to enable data systems to see and hear unstructured data for unified type-agnostic analytics.

Other Collaborative Work

  • Understanding and Benchmarking the Impact of GDPR on Database Systems
    Supreeth Shastri, Vinay Banakar, Melissa Wasserman, Arun Kumar, and Vijay Chidambaram
    VLDB 2020 | Paper PDF | TechReport | Webpage | Talk videos: Youtube Bilibili

  • Hierarchical and Distributed Machine Learning Inference Beyond the Edge
    Anthony Thomas, Yunhui Guo, Yeseong Kim, Baris Aksanli, Arun Kumar and Tajana Rosing
    IEEE ICNSC 2019 | Paper PDF

  • Predicting Eating Events in Free Living Individuals
    Jiayi Wang, Jiue-An Yang, Supun Nakandala, Arun Kumar and Marta M. Jankowska
    eScience 2019 Conference (Poster)

  • In-RDBMS Hardware Acceleration of Advanced Analytics
    Divya Mahajan, Joon Kyung Kim, Jacob Sacks, Adel Ardalan, Arun Kumar, and Hadi Esmaeilzadeh
    VLDB 2018 | Paper PDF | Addendum

Past Projects


Project Orion
Enabling and optimizing ML “over” joins; the precursor to Project Mopheus.


Project Columbus
Enabling and optimizing declarative exploratory feature selection over structured data.


Project Bismarck
Devising a unified system architecture for in-RDBMS implementations of ML algorithms.


Project Staccato
Integrating OCR data with an RDBMS and trading off accuracy for runtime with approximate inference.