Scholar
Jesse Hoogland
Google Scholar ID: KzeLSKMAAAAJ
Executive Director, Timaeus
Citations & Impact
Citations
201
H-index
4
i10-index
4
Publications
13
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- - Paper: Influence Dynamics and Stagewise Data Attribution (2025-10-14)
- - Paper: Compressibility Measures Complexity: Minimum Description Length Meets Singular Learning Theory (2025-10-14)
- - Paper: The Loss Kernel: A Geometric Probe for Deep Learning Interpretability (2025-10-01)
- - Talk: Singular Learning Theory & AI Safety (2025-06-26)
- - Talk: Embryology of AI (2025-06-19)
- - Talk: Jesse Hoogland on Singular Learning Theory (2024-12-01)
Research Experience
- Engaged in AI safety-related research within the Timaeus team, including but not limited to:
- - Influence Dynamics and Stagewise Data Attribution
- - Compressibility Measures Complexity: Minimum Description Length Meets Singular Learning Theory
- - The Loss Kernel: A Geometric Probe for Deep Learning Interpretability
Background
- Co-founder and executive director of Timaeus, an AI safety research org working on applications of Singular Learning Theory (SLT) for AI safety. SLT establishes a connection between the geometry of the loss landscape and internal structure in models, which we are using to develop scalable, rigorous tools for evaluating, interpreting, and aligning neural networks.
Miscellany
- Personal interests may include AI safety, Singular Learning Theory, and related mathematical and computer science research.
Co-authors