Curriculum Vitae

Education

• Ph.D. in Computer Science, University of Southern California, 2025-Present
• M.S. in Computer Science, Cornell University, 2022-2024
• B.A in Mathematics and Computer Science (Magna Cum Laude), Cornell University, 2018-2022

Coursework

Algorithms †‡                       Deep Learning †                         Complexity Theory †‡               Combinatorics †‡
Learning Theory †               Network Algorithms †                Pseudorandomness †               Analysis †‡
Control Theory †                  Lattice Algorithms †                  Cryptography                             Algebra ‡

†: graduate coursework, ‡: multiple courses

Publications

• Just how hard are rotations of ℤ^n? Algorithms and cryptography with the simplest lattice
  Huck Bennett, Atul Ganju, Pura Peetathawatchai, and Noah Stephens-Davidowitz
  EUROCRYPT 2023
  [Arxiv]

Manuscripts/Working Manuscripts

• A Theory of Time-Sensitive Language Generation: Sparse Hallucination Beats Mode Collapse
  Atul Ganju, Travis McVoy, Shaddin Dughmi, Shang-Hua Teng
  In Submission
  [Arxiv]

• Active Learning via Regression Beyond Realizability
  Atul Ganju, Shashaank Aiyer, Ved Sriraman, and Karthik Sridharan
  In Submission
  [Arxiv]

Research Experience

• Research Assistant @ University of Southern California (September 2023-May 2025):
  • Equipped a well-studied framework for langauge generation to incorporate a notion of time-sensitivity. Showed that hallucination is necessary to achieve a breadth of output.
  • Exploring the role of auxiliary data, both distributional and functional, in facilitating learning.
  • Developing interactive learning algorithms that are robust to distributional/modeling assumptions weaker than realizability and algorithms that can achieve adaptive regret bounds.
• Research Assistant @ Cornell University (September 2023-May 2025):
  • Designed and analyzed the first stream-based active learning algorithm with provable guarantees when the problem is not realizable. Mentored two undergraduate students in their first research experience in machine learning.
• Undergraduate Research Assistant @ Cornell University (January 2021-February 2022):
  • Designed, implemented, and analyzed the results of experiments that tested the performance of basis reduction algorithms on bases for rotations of the integer lattice generated from a variety of basis sampling algorithms. Experiments were implemented in Python.
  • Designed, implemented, and analyzed the results of experiments that determined how heuristic sieving algorithms perform on the integer lattice. Experiments were implemented in C++.
  • Proved existence of orthogonal projection-based Cook reduction from ZSVP to 2-SVP.

Teaching Experience

• Graduate Teaching Assistant @ Cornell University
  • CS 6784 Special Topics in Machine Learning: Control Theory (Fall 2022)
  • CS 4820: Introduction to Analysis of Algorithms (Spring 2023)
  • CS 4783: Mathematical Foundations for Machine Learning (Fall 2023)
  • CS 4780: Introduction to Machine Learning (Spring 2024)
• Head Teaching Assistant @ Cornell University
  • CS 4820: Introduction to Analysis of Algorithms (Summer 2021, Spring 2022)
  • CS 4780: Introduction to Machine Learning (Fall 2021)
• Teaching Assistant @ Cornell University
  • CS 4820: Introduction to Analysis of Algorithms (Spring 2021)
  • CS 4780: Introduction to Machine Learning (Spring 2020, Fall 2020)

Technical Skills

• Languages: Python, Java, C, C++, Mathematica, OCaml, MATLAB, SQL
• Technologies/Libraries: Github, Pytorch, Tensorflow, Jupyter