Todd Morrill
I’m a PhD student in Computer Science at Columbia University, advised by Richard Zemel and a member of the Columbia Core AI Lab (CAIL). I’m also a visiting scholar in Anthony Zador’s lab at Cold Spring Harbor Laboratory, and I lead the Biological Learning Working Group at ARNI, the NSF AI Institute for Artificial and Natural Intelligence.
I’m broadly interested in machine learning and neuroscience. My research focuses on memory models, biologically plausible learning and credit assignment, and unsupervised multimodal representation learning. My current research vision is to develop learning algorithms that are amenable to future hardware accelerators, with an eye towards systems that are energy-efficient, online, and continually learning. Before starting my PhD, I completed an MS in Computer Science at Columbia, advised by Richard Zemel and Kathleen McKeown, and spent ten years as a machine learning engineer at PwC.
News
- Jul 2026Presented Bullet Trains: Parallelizing Training of Temporally Precise Spiking Neural Networks at ICML 2026 in Seoul, Korea.
- Jan 2026Teaching assistant for Continual Learning & Memory Models at Columbia, taught by Richard Zemel.
- Dec 2025Presented mixture-of-experts models for survival analysis at NeurIPS 2025 in San Diego, CA, at both the Machine Learning for Health Symposium and the Learning from Time Series for Health workshop.
- Sep 2024Started my PhD in Computer Science at Columbia University, advised by Richard Zemel.
- May 2024Presented Social Orientation: A New Feature for Dialogue Analysis at LREC-COLING 2024 in Turin, Italy, and Prompt Risk Control at ICLR 2024 in Vienna, Austria.
- May 2024Joined Anthony Zador's lab at Cold Spring Harbor Laboratory as a NeuroAI research intern.
Research
@inproceedings{morrill2026bullettrains,
title = {Bullet Trains: Parallelizing Training of Temporally Precise Spiking Neural Networks},
author = {Morrill, Todd and Pehle, Christian and Zador, Anthony},
booktitle = {Proceedings of the 43rd International Conference on Machine Learning (ICML)},
year = {2026},
url = {https://arxiv.org/abs/2603.13283}
}
@inproceedings{morrill2025experts,
title = {Let the Experts Speak: Improving Survival Prediction \& Calibration via Mixture-of-Experts Heads},
author = {Morrill, Todd and Puli, Aahlad Manas and Megjhani, Murad and Park, Soojin and Zemel, Richard},
booktitle = {Proceedings of the Machine Learning for Health (ML4H) Symposium},
year = {2025},
url = {https://arxiv.org/abs/2511.09567}
}
@inproceedings{morrill2024social,
title = {Social Orientation: A New Feature for Dialogue Analysis},
author = {Morrill, Todd and Deng, Zhaoyuan and Chen, Yanda and Ananthram, Amith and Leach, Colin Wayne and McKeown, Kathleen},
booktitle = {Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)},
year = {2024},
url = {https://arxiv.org/abs/2403.04770}
}
@inproceedings{zollo2024prompt,
title = {Prompt Risk Control: A Rigorous Framework for Responsible Deployment of Large Language Models},
author = {Zollo, Thomas P. and Morrill, Todd and Deng, Zhun and Snell, Jake C. and Pitassi, Toniann and Zemel, Richard},
booktitle = {The Twelfth International Conference on Learning Representations (ICLR)},
year = {2024},
url = {https://arxiv.org/abs/2311.13628}
}
Patents and technical reports are listed in my CV.
Beyond research
I live in New York City with my wife Emily and our sheepadoodle, Ziggy. Outside of work, I like to bike, hike, camp, ski, and cook. In a former life, I loved to study languages and still speak Mandarin Chinese well enough.
Recent posts
Chatting with Yourself
TLDR; I pointed a local LLM at my Obsidian vault and asked questions about my notes.
Unsupervised Representation Learning with Predictive Coding
TLDR; Here’s a vanilla implementation of unsupervised predictive coding in PyTorch.
Yet Another Variational Autoencoders Tutorial
TLDR; I’m working through the details of VAEs once and for all.