Nick Watters
PhD Candidate, Department of Brain and Cognitive Sciences, MIT
Curriculum Vitae
Google Scholar
LinkedIn

About Me

I am a PhD candidate in the Department of Brain and Cognitive Sciences at MIT. I work with Mehrdad Jazayeri and Josh Tenenbaum studying how the primate brain represents multi-object scenes, predicts the dynamics of moving objects, and flexibly executes manual grasping and insertion of objects. More broadly, I'm interested in studying neuroscientific questions that bear on shortcomings of current AI, such as how the brain does sample-efficient learning, flexible generalization, and robust motor control. Going forward I plan to work on brain computer interfaces and robotics.

Before beginning my PhD I worked for three years as a research engineer at Google Deepmind, where I focused on visual unsupervised learning and reinforcement learning. Before that I did my undergrad at Harvard, where I majored in math and computer science.

Aside from research, I like playing music (jazz piano) and sports (ultimate frisbee, squash, soccer, hiking).

Ongoing Projects

Representation of Multi-Object Scenes in Primate Frontal Cortex
COSYNE 2023, Work in progress
Nick Watters, Jack Gabel, Joshua Tenenbaum, Mehrdad Jazayeri
Neural Mechanisms of Kinematics Prediction in the Primate Brain
Society for Neuroscience 2023, Work in progress
Nick Watters, Jack Gabel, Joshua Tenenbaum, Mehrdad Jazayeri

Selected Publications

Modeling Human Eye Movements with Neural Networks in a Maze-Solving Task
arXiv 2022, PMLR 2022, CCN 2022
Jason Li, Nick Watters, Yingting Wang, Hansem Sohn, Mehrdad Jazayeri
Modular Object-Oriented Games
arXiv 2021
Nick Watters, Joshua Tenenbaum, Mehrdad Jazayeri
COBRA: Data-Efficient Model-Based RL through Unsupervised Object Discovery and Curiosity-Driven Exploration
arXiv 2019, ICML Workshop 2019
Nick Watters, Loic Matthey, Matko Bosnjak, Christopher P. Burgess, Alexander Lerchner
Spriteworld: A Flexible, Configurable Reinforcement Learning Environment
GitHub 2019
Nick Watters, Loic Matthey, Sebastian Borgeaud, Rishabh Kabra, Alexander Lerchner
Multi-Object Representation Learning with Iterative Variational Inference
ICML 2019
Klaus Greff, Raphaƫl Lopez Kaufman, Rishabh Kabra, Nick Watters, Chris Burgess, Daniel Zoran, Loic Matthey, Matthew Botvinick, Alexander Lerchner
MONet: Unsupervised Scene Decomposition and Representation
arXiv 2019
Christopher Burgess, Loic Matthey, Nick Watters, Rishabh Kabra, Irina Higgins, Matt Botvinick, Alexander Lerchner
Spatial Broadcast Decoder: A Simple Architecture for Learning Disentangled Representations in VAEs
arXiv 2019, ICLR LLD workshop 2019
Nick Watters, Loic Matthey, Christopher P. Burgess, Alexander Lerchner
Visual Interaction Networks
NeurIPS 2017
Nick Watters, Daniel Zoran, Theophane Weber, Peter Battaglia, Razvan Pascanu, Andrea Tacchetti
Neuronal Spike Train Entropy Estimation by History Clustering
Neural Computation 2014
Nick Watters, George Reeke