Follow
Ryan Theisen
Ryan Theisen
Verified email at berkeley.edu
Title
Cited by
Cited by
Year
Taxonomizing local versus global structure in neural network loss landscapes
Y Yang, L Hodgkinson, R Theisen, J Zou, JE Gonzalez, K Ramchandran, ...
Advances in Neural Information Processing Systems 34, 18722-18733, 2021
212021
Asymptotic flocking for the three-zone model
F Cao, S Motsch, A Reamy, R Theisen
Mathematical Biosciences and Engineering 17 (6), 7692-7707, 2020
132020
Deterministic versus stochastic consensus dynamics on graphs
D Weber, R Theisen, S Motsch
Journal of Statistical Physics 176, 40-68, 2019
102019
Evaluating natural language processing models with generalization metrics that do not need access to any training or testing data
Y Yang, R Theisen, L Hodgkinson, JE Gonzalez, K Ramchandran, ...
arXiv preprint arXiv:2202.02842, 2022
82022
Good classifiers are abundant in the interpolating regime
R Theisen, J Klusowski, M Mahoney
International Conference on Artificial Intelligence and Statistics, 3376-3384, 2021
7*2021
Evaluating State-of-the-Art Classification Models Against Bayes Optimality
R Theisen, H Wang, LR Varshney, C Xiong, R Socher
Advances in Neural Information Processing Systems 34, 2021
52021
Global capacity measures for deep relu networks via path sampling
R Theisen, JM Klusowski, H Wang, NS Keskar, C Xiong, R Socher
arXiv preprint arXiv:1910.10245, 2019
42019
When are ensembles really effective?
R Theisen, H Kim, Y Yang, L Hodgkinson, MW Mahoney
Advances in Neural Information Processing Systems 36, 2024
32024
Preference Optimization for Molecular Language Models
R Park, R Theisen, N Sahni, M Patek, A Cichońska, R Rahman
arXiv preprint arXiv:2310.12304, 2023
32023
Test accuracy vs. generalization gap: Model selection in nlp without accessing training or testing data
Y Yang, R Theisen, L Hodgkinson, JE Gonzalez, K Ramchandran, ...
Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and …, 2023
22023
Beyond Worst-Case Generalization in Modern Machine Learning
RC Theisen
University of California, Berkeley, 2023
12023
Leveraging multiple data types for improved compound-kinase bioactivity prediction
R Theisen, T Wang, B Ravikumar, R Rahman, A Cichońska
bioRxiv, 2024.03. 07.583951, 2024
2024
The system can't perform the operation now. Try again later.
Articles 1–12