Hi, I’m Nathan! I’m a PhD student at MIT Brain and Cognitive Sciences, where I study rapid learning and systematic generalization in brains and machines.

Before starting my PhD, I completed a master’s in Applied Mathematics and Physics at UCLouvain in Belgium. I worked as a research assistant in Robert Yang’s lab at MIT, developing scalable methods to evaluate Recurrent Neural Network models. I also had the opportunity to work in Omri Barak’s lab at the Technion, studying generalization in computational models of neural dynamics.

Recently, I’ve been exploring how Large Language Models (LLMs) can be used for writing scientific code. I’m particularly excited about their potential to transform computational modeling, making it significantly more efficient for scientists to explore the space of possible models.

Publications

• Generating and Validating Agent and Environment Code for Simulating Realistic Personality Profiles with Large Language Models.
  Nathan Cloos, M Ganesh Kumar, Adam Manoogian, Christopher J. Cueva^†, Shawn A. Rhoads^†.
  NeurIPS 2024 Workshop on Behavioral ML. [pdf]
• A Framework for Standardizing Similarity Measures in a Rapidly Evolving Field.
  Nathan Cloos, Guangyu Robert Yang, Christopher J. Cueva.
  NeurIPS 2024 Workshop UniReps. [pdf] [code]
• Baba Is AI: Break the Rules to Beat the Benchmark.
  Nathan Cloos, Meagan Jens, Michelangelo Naim, Yen‑Ling Kuo, Ignacio Cases, Andrei Barbu^†, Christopher J. Cueva^†.
  ICML 2024 Workshop on LLMs and Cognition. [pdf] [code]
• Differentiable optimization of similarity scores between models and brains.
  Nathan Cloos, Moufan Li, Markus Siegel, Scott L. Brincat, Earl K. Miller, Guangyu Robert Yang, Christopher J. Cueva.
  ICLR 2024 Workshop on Representational Alignment. [pdf] [code] [website]
• Scaling up the Evaluation of Recurrent Neural Network Models for Cognitive Neuroscience.
  Nathan Cloos, Moufan Li, Guangyu Robert Yang, Christopher J. Cueva.
  Cognitive Computational Neuroscience 2022. [pdf]

^† Equal senior authors