Long Pham

Long Pham

PhD student in Computer Science

Carnegie Mellon University

Biography

I am a final-year PhD student in Computer Science at Carnegie Mellon University (CMU). I am advised by Prof. Jan Hoffmann and collaborate with Prof. Feras Saad. I have been working on type-based resource analysis of programs and probabilistic programming. Currently, I investigate how to incorporate Bayesian inference into type-based resource analysis of programs. I also work on a type system for checking the soundness of programmable inference in probabilistic programming, together with Prof. Di Wang at Peking University.

Before coming to CMU, I obtained a B.A. and an M.Sc. in Computer Science from the University of Oxford. For my B.A., I did a research project on higher-order constrained Horn clauses under the supervision of Prof. Luke Ong and Dr. Steven Ramsay. For my M.Sc., I worked with Prof. Marta Kwiatkowska and Dr. Wenjie Ruan, investigating how the choice of loss functions used during training affects the robustness of deep neural networks.

Interests

  • Resource analysis of programs
  • Static program analysis (e.g., type systems)
  • Probabilistic programming and Bayesian inference
  • Dynamic program analysis (e.g., fuzzing)
  • Concurrent programming and session types

Education

  • Ph.D. in Computer Science, Present

    Carnegie Mellon University

  • M.Sc. in Computer Science, 2019

    University of Oxford

  • B.A. in Computer Science, 2018

    University of Oxford

News

January 2025   Thesis proposal at CMU

October 2024   Presented Programmable MCMC with Soundly Composed Guide Programs at OOPSLA 2024 in Pasadena

June 2024   Presented Robust Resource Bounds with Static Analysis and Bayesian Inference at PLDI 2024 in Copenhagen

March 2024   Wrote a blog post about my research on hybrid resource analysis

June - August 2021   Research internship at Automated Reasoning Group (ARG) of Amazon Web Services (AWS) in Boston

September 2020   Our submission to CSL 2021 was accepted.

July 2020   This personal website is created.

Publications

Thesis Proposal: Hybrid Resource-Bound Analysis of Programs

Resource-bound analysis aims to infer symbolic bounds of worst-case resource usage (e.g., running time, memory, and energy) of programs …

Programmable MCMC with Soundly Composed Guide Programs (OOPSLA 2024)

Probabilistic programming languages (PPLs) provide language support for expressing flexible probabilistic models and solving Bayesian …

Robust Resource Bounds with Static Analysis and Bayesian Inference (PLDI 2024)

There are two approaches to automatically deriving symbolic worst-case resource bounds for programs: static analysis of the source code …

Worst-Case Input Generation for Concurrent Programs Under Non-Monotone Resource Metrics (LMCS 2024)

Worst-case input generation aims to automatically generate inputs that exhibit the worst-case performance of programs. It has several …

Typable Fragments of Polynomial Automatic Amortized Resource Analysis (CSL 2021)

Being a fully automated technique for resource analysis, automatic amortized resource analysis (AARA) can fail in returning worst-case …

Contact

  • longp@andrew.cmu.edu
  • Computer Science Department, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213
  • My office at CMU is GHC 5113