Scholar
Pranay Mundra
Google Scholar ID: 7IJcHDwAAAAJ
Yale University
Citations & Impact
Citations
32
H-index
3
i10-index
1
Publications
5
Co-authors
13
list available
Resume (English only)
Academic Achievements
- Developed KOIOS: a novel filter verification system for top-k semantic similarity set search
- Contributed to Quok: a system for approximate query answering over Open Knowledge
- Proposed Fair K Coverage (FKC) coresets for machine learning, preserving fairness and coverage while selecting smaller representative data subsets
- Built Maimon framework: for approximating acyclic schema discovery from relations using Multivalued Dependencies
- Worked on LightDB and other Visual Database Management Systems
- Contributed to the development of Tree Stack Memory Units, a novel neural network architecture
Research Experience
- Currently a Ph.D. student at Yale CS, researching practical differential privacy in distributed graph algorithms
- Previously worked as a Research Data Engineer II at the University of Rochester Medical Center, Department of Biostatistics & Computational Biology, with the McCall Research Lab
- As a Graduate Research Assistant, developed KOIOS—a novel filter verification system for top-k semantic similarity set search—and contributed to Quok, a system for approximate query answering over Open Knowledge
- Visiting researcher at Paris Lodron Universität Salzburg Database Research Group under Professor Martin Schäler, co-developing an alignment algorithm for uncovering semantic similarities in multilingual biblical texts
- Visiting researcher at MIT CSAIL Parallel Computing Group with Professors Quanquan Liu and Julian Shun, working on a benchmark suite for distributed privacy-preserving locally adjustable graph algorithms
- Research intern at Caltech Tensorlab, investigating neural network architectures’ ability to understand compositional data structures and helping develop Tree Stack Memory Units
Background
- Research interests include Database Systems, Machine Learning, and Differential Privacy
- Focuses on data discovery and mining problems such as minimizing data bias, improving data quality for machine learning, differential privacy systems, and query models
- Current research involves selecting coresets for large datasets to train ML models with equivalent accuracy while reducing computational costs and improving speed
- Also developing solutions for fast aggregate queries over massive knowledge graphs with missing values
- Enjoys working on problems in number theory, group theory, probability, combinatorics, and graph theory with applications in computer science
