Sotabase

Career

· Research Scientist, Google2024–
· Research Intern, Google Research2024–2024
· Applied Sciences Intern, Microsoft Research2023–2023
· Research Intern (PhD), Allen Institute for AI2022–2022
· PhD in Computer Science, Columbia University2019–2024
· Associate Data Scientist, LinkedIn2018–2019
· Sc.B. with Honors in Applied Math - Computer Science, Brown University2014–2018

Publications (17)

Representational Strengths and Limitations of Transformers
Neural Information Processing Systems · 2023
123
cited
Learning Single-Index Models with Shallow Neural Networks
Neural Information Processing Systems · 2022
92
cited
Transformers, parallel computation, and logarithmic depth
International Conference on Machine Learning · 2024
58
cited
Support vector machines and linear regression coincide with very high-dimensional features
Neural Information Processing Systems · 2021
31
cited
On the Approximation Power of Two-Layer Networks of Random ReLUs
Annual Conference Computational Learning Theory · 2021
29
cited
One-layer transformers fail to solve the induction heads task
arXiv.org · 2024
26
cited
Enabling Equation-Free Modeling via Diffusion Maps
Journal of Dynamics and Differential Equations · 2021
11
cited
Intrinsic dimensionality and generalization properties of the R-norm inductive bias
Annual Conference Computational Learning Theory · 2022
7
cited
Near-Optimal Statistical Query Lower Bounds for Agnostically Learning Intersections of Halfspaces with Gaussian Marginals
Annual Conference Computational Learning Theory · 2022
6
cited
Improving the predictions of ML-corrected climate models with novelty detection
2022
5
cited
Spliceman2: a computational web server that predicts defects in pre-mRNA splicing
Bioinform. · 2017
5
cited
On Scrambling Phenomena for Randomly Initialized Recurrent Networks
Neural Information Processing Systems · 2022
2
cited
Expressivity of Neural Networks via Chaotic Itineraries beyond Sharkovsky's Theorem
International Conference on Artificial Intelligence and Statistics · 2021
1
cited
Applying Rademacher-Like Bounds to Combinatorial Samples and Function Selection
2018
Fast attention mechanisms: a tale of parallelism
arXiv.org · 2025
T HE E FFECT OF M ODEL C APACITY ON THE E MER - GENCE OF I N -C ONTEXT L EARNING IN T RANSFORMERS
Uniform Convergence Bounds for Codec Selection
arXiv.org · 2018
Sotabase