Andy Shih

Google Scholar / Github / Twitter

I received my PhD in Computer Science from Stanford University in 2024, where I was co-advised by Stefano Ermon and Dorsa Sadigh. I was a part of the Stanford Artificial Intelligence Laboratory (SAIL) and Stanford Machine Learning Group.

I got my BS and MS in Computer Science from UCLA, where I worked in the Automated Reasoning Group with Arthur Choi and Adnan Darwiche.

Here is my CV.

Selected Publications [Full List]

  • Parallel Sampling of Diffusion Models
    Andy Shih and Suneel Belkhale and Stefano Ermon and Dorsa Sadigh and Nima Anari
    NeurIPS 2023 — 37th Conference on Neural Information Processing Systems, New Orleans, USA, December 2023.
    [bib] [arxiv] [slides] [poster] [code]
    Spotlight Presentation

  • Long Horizon Temperature Scaling
    Andy Shih and Dorsa Sadigh and Stefano Ermon
    ICML 2023 — 40th International Conference on Machine Learning, Hawaii, USA, July 2023.
    [bib] [arxiv] [pdf] [slides] [poster] [code]

  • Training and Inference on Any-Order Autoregressive Models the Right Way
    Andy Shih and Dorsa Sadigh and Stefano Ermon
    NeurIPS 2022 — 36th Conference on Neural Information Processing Systems, New Orleans, USA, December 2022.
    [bib] [arxiv] [pdf] [slides] [poster] [code]
    Oral Presentation [top 1.9%]
    2023 UAI Workshop on Tractable Probabilistic Modeling Best Paper Honorable Mention

  • Imitation Learning by Estimating Expertise of Demonstrators
    Mark Beliaev* and Andy Shih* and Stefano Ermon and Dorsa Sadigh and Ramtin Pedarsani
    ICML 2022 — 39th International Conference on Machine Learning, Baltimore, USA, July 2022.
    [bib] [arxiv] [pdf] [code]

  • On the Critical Role of Conventions in Adaptive Human-AI Collaboration
    Andy Shih and Arjun Sawhney and Jovana Kondic and Stefano Ermon and Dorsa Sadigh
    ICLR 2021 — 9th International Conference on Learning Representations, Virtual, May 2021.
    [bib] [arxiv] [pdf] [slides] [poster] [code] [blogpost]

  • Probabilistic Circuits for Variational Inference in Discrete Graphical Models
    Andy Shih and Stefano Ermon
    NeurIPS 2020 — 34th Conference on Neural Information Processing Systems, Vancouver, Canada, December 2020.
    [bib] [arxiv] [pdf] [slides] [poster] [code] [blogpost]

  • Smoothing Structured Decomposable Circuits
    Andy Shih and Guy Van den Broeck and Paul Beame and Antoine Amarilli
    NeurIPS 2019 — 33rd Conference on Neural Information Processing Systems, Vancouver, Canada, December 2019.
    [bib] [arxiv] [pdf] [slides] [poster] [code]
    Spotlight Presentation [top 2.4%]



Awards

  • 2023 NeurIPS Spotlight Presentation
  • 2023 UAI Workshop on Tractable Probabilistic Modeling Best Paper Honorable Mention
  • 2022 NeurIPS Top Reviewer Award
  • 2022 NeurIPS Oral Presentation (top 1.9%)
  • 2019 NeurIPS Spotlight Presentation (top 2.4%)
  • International Collegiate Programming Contest (ICPC) World Finals (x2)
  • 2019 UCLA Computer Science Outstanding Master's Student Award


Service

Reviewer: ICML (2020, 2021, 2022, 2023), NeurIPS (2020, 2021, 2022, 2023), ICLR (2021, 2022, 2023), CoRL (2020, 2021), AISTATS (2022), AAAI (2022)

Coach: Stanford ACM-ICPC (2020-2023)

Contact me at

andyshih at cs dot stanford dot edu