Abolfazl Asudeh 🔊

Associate Professor
Department of Computer Science
University of Illinois Chicago
InDeX Lab

Contact
851 S. Morgan St., 11th Floor SEO, Room 1131, Chicago, IL 60607
☏: (312) 413-8028 ✉: asudeh[AT]uic[DOT]edu
A. Asudeh is an Associate Professor of Computer Science at the University of Illinois Chicago and the director of Innovative Data Exploration Laboratory (InDeX Lab). He is a senior member of ACM and a senior member of IEEE. He serves as an Associate Editor for the IEEE Transactions on Knowledge and Data Engineering (TKDE), a VLDB Ambassador, and the VLDB Endowment's Liaison to NSF.
His research focus is on Algorithm Design for Data and AI problems. He designs efficient, accurate, and responsible solutions that leverage Approximation, Randomized, and Computational Geometry Algorithms. His research is supported NSF and has received recognitions, including the Communications of the ACM's (CACM) Research Highlight, Google's Research Scholar Award, SIGMOD 2019 Research Highlight Award, “Best of VLDB” 2020 (the special issue of VLDBJ), and SIGMOD 2017 Reproducibility Award.
 
Recent Publications (Since 2024) -- see publications for the complete list.
 
📌 Pinned Systems and Repositories 📌

Needle🪡🔍 is a deployment-ready open-source image retrieval database with high accuracy that can handle complex queries in natural language. It is Fast, Efficient, and Precise, outperforming state-of-the-art methods. Born from high-end research, Needle is designed to be accessible to everyone while delivering top-notch performance. Whether you are a researcher, developer, or an enthusiast, Needle opens up innovative ways to explore your image datasets. ✨

📖 Detailed installation instructions: Getting Started .


RSR 🧮: Efficient Matrix Multiplication for Accelerating Inference in Binary and Ternary Neural Networks
This project aims to provide a fast and efficient approach to low-bit matrix multiplication. The code repository implements Redundant Segment Reduction (RSR), a fast matrix multiplication algorithm designed for matrices in binary and ternary networks. The RSR method optimizes computation efficiency by a log(n) factor, making it particularly useful for applications in low-bit deep learning and efficient inference. The codebase provides ready-to-use C++ and NumPy-based implementations, as well as PyTorch implementations with both CPU and GPU support, enabling scalable and optimized matrix operations in deep learning environments. It includes sample experiments on various `1.58bit` models and LLMs.✨
 
Sponsors
  • NSF IIS-2348919 (2024 - 2027): III: Small: Fairness-aware Data Structures for Approximate Query Processing. Abolfazl Asudeh and Stavros Sintos.
  • NSF IIS-2107290 (2021 - 2024): III: Medium: Collaborative Research: Fairness in Web Database Applications. Abolfazl Asudeh (Lead PI - UIC), H. V. Jagadish (UofM), and Gautam Das and Shirin Nilizadeh (UTA).
  • Google Research Scholar Award (2021 - 2022): An end-to-end system for detecting cherry-picked trendlines. Abolfazl Asudeh.

National Science Foundation Google Research CloudBank