Seguir
Blake Bordelon
Blake Bordelon
Applied Mathematics at Harvard
Dirección de correo verificada de g.harvard.edu
Título
Citado por
Citado por
Año
Spectrum dependent learning curves in kernel regression and wide neural networks
B Bordelon, A Canatar, C Pehlevan
International Conference on Machine Learning, 1024-1034, 2020
1622020
Spectral bias and task-model alignment explain generalization in kernel regression and infinitely wide neural networks
A Canatar, B Bordelon, C Pehlevan
Nature communications 12 (1), 2914, 2021
1482021
Neural networks as kernel learners: The silent alignment effect
A Atanasov, B Bordelon, C Pehlevan
arXiv preprint arXiv:2111.00034, 2021
542021
Self-consistent dynamical field theory of kernel evolution in wide neural networks
B Bordelon, C Pehlevan
Advances in Neural Information Processing Systems 35, 32240-32256, 2022
402022
Dispersive optical model analysis of generating a neutron-skin prediction beyond the mean field
MC Atkinson, MH Mahzoon, MA Keim, BA Bordelon, CD Pruitt, RJ Charity, ...
Physical Review C 101 (4), 044303, 2020
312020
Population codes enable learning from few examples by shaping inductive bias
B Bordelon, C Pehlevan
Elife 11, e78606, 2022
172022
The influence of learning rule on representation dynamics in wide neural networks
B Bordelon, C Pehlevan
The Eleventh International Conference on Learning Representations, 2022
162022
Out-of-distribution generalization in kernel regression
A Canatar, B Bordelon, C Pehlevan
Advances in Neural Information Processing Systems 34, 12600-12612, 2021
152021
Learning Curves for SGD on Structured Features
B Bordelon, C Pehlevan
International Conference on Learning Representations, 2022
142022
A theory of neural tangent kernel alignment and its influence on training
H Shan, B Bordelon
arXiv preprint arXiv:2105.14301, 2021
14*2021
The onset of variance-limited behavior for networks in the lazy and rich regimes
A Atanasov, B Bordelon, S Sainathan, C Pehlevan
arXiv preprint arXiv:2212.12147, 2022
132022
Dynamics of finite width kernel and prediction fluctuations in mean field neural networks
B Bordelon, C Pehlevan
Advances in Neural Information Processing Systems 36, 2024
102024
Feature-learning networks are consistent across widths at realistic scales
N Vyas, A Atanasov, B Bordelon, D Morwani, S Sainathan, C Pehlevan
Advances in Neural Information Processing Systems 36, 2024
92024
Grokking as the transition from lazy to rich training dynamics
T Kumar, B Bordelon, SJ Gershman, C Pehlevan
arXiv preprint arXiv:2310.06110, 2023
52023
Depthwise hyperparameter transfer in residual networks: Dynamics and scaling limit
B Bordelon, L Noci, MB Li, B Hanin, C Pehlevan
arXiv preprint arXiv:2309.16620, 2023
42023
Efficient online inference for nonparametric mixture models
R Schaeffer, B Bordelon, M Khona, W Pan, IR Fiete
Uncertainty in Artificial Intelligence, 2072-2081, 2021
42021
Self-consistent dynamical field theory of kernel evolution in wide neural networks
B Bordelon, C Pehlevan
Journal of Statistical Mechanics: Theory and Experiment 2023 (11), 114009, 2023
32023
Capacity of Group-invariant Linear Readouts from Equivariant Representations: How Many Objects can be Linearly Classified Under All Possible Views?
M Farrell, B Bordelon, S Trivedi, C Pehlevan
arXiv preprint arXiv:2110.07472, 2021
22021
Loss dynamics of temporal difference reinforcement learning
B Bordelon, P Masset, H Kuo, C Pehlevan
Advances in Neural Information Processing Systems 36, 2024
12024
A Dynamical Model of Neural Scaling Laws
B Bordelon, A Atanasov, C Pehlevan
arXiv preprint arXiv:2402.01092, 2024
12024
El sistema no puede realizar la operación en estos momentos. Inténtalo de nuevo más tarde.
Artículos 1–20