Følg
David Rolnick
David Rolnick
McGill University, Mila Quebec AI Institute
Verifisert e-postadresse på cs.mcgill.ca - Startside
Tittel
Sitert av
Sitert av
År
Why does deep and cheap learning work so well?
HW Lin, M Tegmark, D Rolnick
Journal of Statistical Physics 168, 1223-1247, 2017
6822017
Tackling climate change with machine learning
D Rolnick, PL Donti, LH Kaack, K Kochanski, A Lacoste, K Sankaran, ...
ACM Computing Surveys (CSUR) 55 (2), 1-96, 2022
663*2022
Experience replay for continual learning
D Rolnick, A Ahuja, J Schwarz, TP Lillicrap, G Wayne
Advances in Neural Information Processing Systems, 2019
5502019
Deep learning is robust to massive label noise
D Rolnick, A Veit, S Belongie, N Shavit
arXiv preprint arXiv:1705.10694, 2017
5382017
How to start training: The effect of initialization and architecture
B Hanin, D Rolnick
Advances in Neural Information Processing Systems, 2018
2172018
The power of deeper networks for expressing natural functions
D Rolnick, M Tegmark
International Conference on Learning Representations, 2018
1942018
Complexity of linear regions in deep networks
B Hanin, D Rolnick
International Conference on Machine Learning, 2596-2604, 2019
1612019
Deep ReLU networks have surprisingly few activation patterns
B Hanin, D Rolnick
Advances in Neural Information Processing Systems, 361-370, 2019
1582019
Measuring and regularizing networks in function space
AS Benjamin, D Rolnick, K Kording
International Conference on Learning Representations, 2019
1012019
DC3: A learning method for optimization with hard constraints
PL Donti, D Rolnick, JZ Kolter
International Conference on Learning Representations, 2021
742021
Reverse-engineering deep ReLU networks
D Rolnick, K Kording
International Conference on Machine Learning, 8178-8187, 2020
69*2020
Aligning artificial intelligence with climate change mitigation
LH Kaack, PL Donti, E Strubell, G Kamiya, F Creutzig, D Rolnick
Nature Climate Change 12 (6), 518-527, 2022
452022
A multi-pass approach to large-scale connectomics
Y Meirovitch, A Matveev, H Saribekyan, D Budden, D Rolnick, G Odor, ...
arXiv preprint arXiv:1612.02120, 2016
402016
Cross-classification clustering: An efficient multi-object tracking technique for 3-D instance segmentation in connectomics
Y Meirovitch, L Mi, H Saribekyan, A Matveev, D Rolnick, N Shavit
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern …, 2019
332019
Morphological error detection in 3D segmentations
D Rolnick, Y Meirovitch, T Parag, H Pfister, V Jain, JW Lichtman, ...
arXiv preprint arXiv:1705.10882, 2017
242017
Graph-coloring ideals: Nullstellensatz certificates, Gröbner bases for chordal graphs, and hardness of Gröbner bases
JA De Loera, S Margulies, M Pernpeintner, E Riedl, D Rolnick, G Spencer, ...
Proceedings of the 2015 ACM on International Symposium on Symbolic and …, 2015
18*2015
Quantitative Tverberg, Helly, & Carathéodory theorems
JA De Loera, RN La Haye, D Rolnick, P Soberón
arXiv preprint arXiv:1503.06116, 2015
16*2015
Bugs in the data: How ImageNet misrepresents biodiversity
AS Luccioni, D Rolnick
AAAI Conference on Artificial Intelligence, 2022
142022
Quantitative (p, q) theorems in combinatorial geometry
D Rolnick, P Soberón
Discrete Mathematics 340 (10), 2516-2527, 2017
14*2017
Quantitative combinatorial geometry for continuous parameters
JA De Loera, RN La Haye, D Rolnick, P Soberón
Discrete & Computational Geometry 57, 318-334, 2017
142017
Systemet kan ikke utføre handlingen. Prøv på nytt senere.
Artikler 1–20