About Me

I am a postdoctoral researcher working with Thomas Griffiths at the Department of Computer Science, Princeton University. I am supported by a DataX postdoctoral fellowship. I received my Ph.D. in Computer Science in 2021 from the University of Toronto under the supervision of Richard Zemel, where I also completed a postdoctoral fellowship.

My recent research interests include nonparametric Bayesian approaches to few-shot and continual learning. I am also interested in uncertainty quantification for reliable machine learning.

Preprints

• A Metalearned Neural Circuit for Nonparametric Bayesian Inference. Jake Snell, Gianluca Bencomo, and Thomas L. Griffiths.

• Few-Shot Attribute Learning.
  Mengye Ren, Eleni Triantafillou, Kuan-Chieh Wang, James Lucas, Jake Snell, Xaq Pitkow, Andreas S. Tolias, and Richard Zemel.

Publications

• Implicit Maximum a Posteriori Filtering via Adaptive Optimization. Gianluca Bencomo, Jake Snell, and Thomas L. Griffiths. ICLR 2024 (to appear).

• Prompt Risk Control: A Rigorous Framework for Responsible Deployment of Large Language Models. Thomas P. Zollo, Todd Morrill, Zhun Deng, Jake Snell, Toniann Pitassi, and Richard Zemel. ICLR 2024 (to appear).

• Distribution-Free Statistical Dispersion Control for Societal Applications. Zhun Deng, Thomas P. Zollo, Jake Snell, Toniann Pitassi, and Richard Zemel. NeurIPS 2023.

• Im-Promptu: In-Context Composition from Image Prompts. Bhishma Dedhia, Michael Chang, Jake Snell, Thomas L. Griffiths, and Niraj Jha. NeurIPS 2023.

• Quantile Risk Control: A Flexible Framework for Bounding the Probability of High-Loss Predictions. Jake Snell, Thomas P. Zollo, Zhun Deng, Toniann Pitassi, and Richard Zemel. ICLR 2023.

• Bayesian Few-Shot Classification with One-vs-Each Pólya-Gamma Augmented Gaussian Processes. Jake Snell and Richard Zemel. ICLR 2021.
  arxiv | code

• Lorentzian Distance Learning for Hyperbolic Representations.
  Marc T. Law, Renjie Liao, Jake Snell, and Richard Zemel. ICML 2019.
  retrieval code | binary classification code | supplemental

• Dimensionality Reduction for Representing the Knowledge of Probabilistic Models.
  Marc T. Law, Jake Snell, Amir-Massoud Farahmand, Raquel Urtasun, and Richard Zemel. ICLR 2019.
  visualization code | flowers zero-shot code | CUB zero-shot code

• Learning Latent Subspaces in Variational Autoencoders.
  Jack Klys, Jake Snell, and Richard Zemel. NeurIPS 2018.

• Meta-Learning for Semi-Supervised Few-Shot Classification.
  Mengye Ren, Eleni Triantafillou, Sachin Ravi, Jake Snell, Kevin Swersky, Josh B. Tenenbaum, Hugo Larochelle, and Richard Zemel. ICLR 2018.
  code

• Prototypical Networks for Few-shot Learning.
  Jake Snell, Kevin Swersky, and Richard Zemel. NeurIPS 2017.
  code | supplemental

• Stochastic Segmentation Trees for Multiple Ground Truths.
  Jake Snell and Richard Zemel. UAI 2017.
  supplemental

• Learning to Generate Images with Perceptual Similarity Metrics.
  Jake Snell, Karl Ridgeway, Renjie Liao, Brett Roads, Michael C. Mozer, and Richard Zemel. ICIP 2017.
  supplemental

Ph.D. Thesis

• Learning to Build Probabilistic Models with Limited Data.
  Jake Snell. Ph.D. Thesis, Dept. of Computer Science, Univ. of Toronto, 2021.

Service

Conference Reviewing

• AutoML-Conf: 2022 (area chair)
• CoLLAs: 2022
• CVPR: 2021
• ICLR: 2020, 2018, 2017
• ICML: 2022, 2019, 2018
• NeurIPS: 2022, 2021 (top 8% reviewer), 2019 (top 6% reviewer), 2018, 2017

Journal Reviewing

• JMLR: 2021
• TMLR: 2022

Workshop Reviewing

• NeurIPS Workshop on Meta-Learning: 2021 (area chair), 2020 (meta-reviewer), 2018, 2017