Joel Phillips Michelson

PhD student, Vanderbilt University

Joined

May 2021

Names

Joel Phillips Michelson (Preferred)

Emails

****@vanderbilt.edu (Confirmed)

Personal Links

Career & Education History

PhD student

Vanderbilt University (vanderbilt.edu)

2017 – Present

Undergrad student

Rhodes College (rhodes.edu)

2013 – 2017

Advisors, Relations & Conflicts

PhD Advisor

Maithilee Kunda

2017 – Present

Expertise

cognitive modeling

2017 – Present

deep learning

2017 – Present

reinforcement learning

2017 – Present

Publications

Extracting Belief-Update Rules to Explain Theory-of-Mind Generalization Failures
Joel Phillips Michelson, Deepayan Sanyal, Maithilee Kunda
- CogInterp @ NeurIPS 2025 Reject
- Readers: Everyone
On the Role of Pretraining in Domain Adaptation in an Infant-Inspired Distribution Shift Task
Deepayan Sanyal, Joel Phillips Michelson, Maithilee Kunda
- CogInterp @ NeurIPS 2025 Poster
- Readers: Everyone
A Neurodiversity-Inspired Solver for the Abstraction & Reasoning Corpus (ARC) Using Visual Imagery and Program Synthesis
James Ainooson, Deepayan Sanyal, Joel Phillips Michelson, Yuan Yang, Maithilee Kunda
- ACS 2025 Oral
- Readers: Everyone
Standoff: benchmarking representation learning for nonverbal theory of mind tasks
Joel Phillips Michelson, Deepayan Sanyal, James Ainooson, Effat Farhana, Maithilee Kunda
- ICDL
- Readers: Everyone
Feature-Based Analysis of Theory of Mind Representations in Neural Network Models
Joel Phillips Michelson, Deepayan Sanyal, Maithilee Kunda
- ICLR 2025 Conference Withdrawn Submission
- Readers: Everyone
Understanding Distribution Alignment Through Category Separability In An Infant-Inspired Domain Adaptation Task
Deepayan Sanyal, Joel Phillips Michelson, Maithilee Kunda
- Submitted to ICLR 2025
- Readers: Everyone
Investigating Feature Alignment Under An Infant-Inspired Visual Distribution Shift
Deepayan Sanyal, James Ainooson, Joel Phillips Michelson, Yuan Yang, Maithilee Kunda
- Submitted to ICLR 2024
- Readers: Everyone

Co-Authors