Wonyeol Lee

PhD Candidate at Stanford CS  |  wonyeol.lee.cs@gmail.com  |  CV

About Me

I am a final year PhD student in Computer Science at Stanford University advised by Alex Aiken. In 2017-2020, I was a researcher at KAIST to serve mandatory military service, working with Hongseok Yang. I received my BS degree in Computer Science & Mathematics from POSTECH. I am a recipient of the Samsung Scholarship.

I like to raise and solve fundamental / theoretical questions in machine learning (ML) and related areas that are connected to practice. In particular, I am interested in identifying and addressing various challenges in ML that arise from discrepancies between the theory and practice of ML, especially the following: in theory, mathematical functions of interest are assumed to be “well-behaved”, but in practice, they are represented by programs which can be easily “ill-behaved”.

Research Interests

• Correctness (verification | program analysis)

• Numerical Programs (probabilistic | differentiable | floating-point)

• Differentiation (non-differentiability | automatic differentiation | gradient estimation)

• Probabilistic Inference (variational inference)

• Deep Learning (low-precision)

Publications

Preprints

1. Training with Mixed-Precision Floating-Point Assignments
   Wonyeol Lee, Rahul Sharma, Alex Aiken
   In Submission

Conferences

2. Smoothness Analysis for Probabilistic Programs with Application to Optimised Variational Inference
   Wonyeol Lee, Xavier Rival, Hongseok Yang
   POPL 2023
   slides | code

3. On Correctness of Automatic Differentiation for Non-Differentiable Functions
   Wonyeol Lee, Hangyeol Yu, Xavier Rival, Hongseok Yang
   NeurIPS 2020 (Spotlight)
   slides | slides-long | poster | video

4. Differentiable Algorithm for Marginalising Changepoints
   Hyoungjin Lim, Gwonsoo Che, Wonyeol Lee, Hongseok Yang
   AAAI 2020

5. Towards Verified Stochastic Variational Inference for Probabilistic Programs
   Wonyeol Lee, Hangyeol Yu, Xavier Rival, Hongseok Yang
   POPL 2020
   slides | video | code

6. Reparameterization Gradient for Non-Differentiable Models
   Wonyeol Lee, Hangyeol Yu, Hongseok Yang
   NeurIPS 2018
   slides | poster | code

7. On Automatically Proving the Correctness of math.h Implementations
   Wonyeol Lee, Rahul Sharma, Alex Aiken
   POPL 2018
   slides | slides-short | video

8. Verifying Bit-Manipulations of Floating-Point
   Wonyeol Lee, Rahul Sharma, Alex Aiken
   PLDI 2016
   slides | video

9. A Proof System for Separation Logic with Magic Wand
   Wonyeol Lee, Sungwoo Park
   POPL 2014

10. CT-IC: Continuously Activated and Time-Restricted Independent Cascade Model for Viral Marketing
    Wonyeol Lee, Jinha Kim, Hwanjo Yu
    ICDM 2012
    slides

11. Edge Detection Using Morphological Amoebas in Noisy Images
    Wonyeol Lee, Seyun Kim, Youngwoo Kim, Jaeyoung Lim, Dong Hoon Lim
    ICIP 2009

Journals

12. CT-IC: Continuously Activated and Time-Restricted Independent Cascade Model for Viral Marketing
    Jinha Kim, Wonyeol Lee, Hwanjo Yu
    Knowledge-Based Systems, 2014

13. Edge Detection Based on Morphological Amoebas
    Wonyeol Lee, Youngwoo Kim, Seyun Kim, Jaeyoung Lim, Dong Hoon Lim
    Imaging Science Journal, 2012

trips | last updated: 01/2023