Sotabase

Career

· Member of Technical Staff - Pretraining Team, Anthropic2022–
· Norman B. Leventhal Presidential Fellow, MIT2010–2010
· Research Assistant, Caltech2005–2017
· BS Engineering and Applied Science, California Institute of Technology (Caltech)
· Senior Developer for Zotero, George Mason University
· Researcher, Google DeepMind
· Researcher, MIT Brain and Cognitive Sciences

Publications (101)

A Simple Framework for Contrastive Learning of Visual Representations
International Conference on Machine Learning · 2020
22,693
cited
Big Self-Supervised Models are Strong Semi-Supervised Learners
Neural Information Processing Systems · 2020
2,454
cited
When Does Label Smoothing Help?
Neural Information Processing Systems · 2019
2,230
cited
Similarity of Neural Network Representations Revisited
International Conference on Machine Learning · 2019
1,825
cited
Do Better ImageNet Models Transfer Better?
Computer Vision and Pattern Recognition · 2018
1,463
cited
Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time
International Conference on Machine Learning · 2022
1,328
cited
Do Vision Transformers See Like Convolutional Neural Networks?
Neural Information Processing Systems · 2021
1,237
cited
Big Self-Supervised Models Advance Medical Image Classification
IEEE International Conference on Computer Vision · 2021
641
cited
OpenFlamingo: An Open-Source Framework for Training Large Autoregressive Vision-Language Models
arXiv.org · 2023
555
cited
Towards Generalist Biomedical AI
NEJM AI · 2023
421
cited
Synthetic Data from Diffusion Models Improves ImageNet Classification
Trans. Mach. Learn. Res. · 2023
398
cited
The Origins and Prevalence of Texture Bias in Convolutional Neural Networks
Neural Information Processing Systems · 2019
317
cited
Do Wide and Deep Networks Learn the Same Things? Uncovering How Neural Network Representations Vary with Width and Depth
International Conference on Learning Representations · 2020
306
cited
Laminar recordings in frontal cortex suggest distinct layers for maintenance and control of working memory
Proceedings of the National Academy of Sciences of the United States of America · 2018
254
cited
Robust and data-efficient generalization of self-supervised machine learning for diagnostic imaging
Nature Biomedical Engineering · 2023
229
cited
Latency and Selectivity of Single Neurons Indicate Hierarchical Processing in the Human Medial Temporal Lobe
Journal of Neuroscience · 2008
228
cited
Boosting Contrastive Self-Supervised Learning with False Negative Cancellation
IEEE Workshop/Winter Conference on Applications of Computer Vision · 2020
211
cited
Patching open-vocabulary models by interpolating weights
Neural Information Processing Systems · 2022
202
cited
A category-specific response to animals in the right human amygdala
Nature Neuroscience · 2011
168
cited
FlexiViT: One Model for All Patch Sizes
Computer Vision and Pattern Recognition · 2022
143
cited
Sotabase
Simon Kornblith | Researcher Profile | Sotabase | Sotabase