Prati
David W. Romero
David W. Romero
Potvrđena adresa e-pošte na vu.nl - Početna stranica
Naslov
Citirano
Citirano
Godina
CKConv: Continuous Kernel Convolution For Sequential Data
DW Romero, A Kuzina, EJ Bekkers, JM Tomczak, M Hoogendoorn
International Conference on Learning Representations (ICLR), 2022, 2021
982021
Attentive Group Equivariant Convolutional Networks
DW Romero, EJ Bekkers, JM Tomczak, M Hoogendoorn
International Conference on Machine Learning (ICML), 2020, 2020
782020
Flexconv: Continuous kernel convolutions with differentiable kernel sizes
DW Romero, RJ Bruintjes, JM Tomczak, EJ Bekkers, M Hoogendoorn, ...
International Conference on Learning Representations (ICLR), 2022, 2021
722021
Group Equivariant Stand-Alone Self-Attention For Vision
DW Romero, JB Cordonnier
International Conference on Learning Representations (ICLR), 2021, 2020
492020
Learning Equivariances and Partial Equivariances from Data
DW Romero, S Lohit
Advances in Neural Information Processing Systems 35 (NeurIPS 2022), 2021
28*2021
Co-Attentive Equivariant Neural Networks: Focusing Equivariance On Transformations Co-Occurring In Data
DW Romero, M Hoogendoorn
International Conference on Learning Representations (ICLR), 2020, 2019
242019
Exploiting Redundancy: Separable Group Convolutional Networks on Lie Groups
DM Knigge, DW Romero, EJ Bekkers
International Conference on Machine Learning (ICML), 2022, 2021
232021
Relaxing Equivariance Constraints with Non-stationary Continuous Filters
TFA van der Ouderaa, DW Romero, M van der Wilk
Advances in Neural Information Processing Systems 36 (NeurIPS 2022), 2022
172022
Wavelet networks: Scale-Translation Equivariant Learning From Raw Time-Series
DW Romero, EJ Bekkers, JM Tomczak, M Hoogendoorn
Transactions of Machine Learning Research, 2023, 2020
16*2020
Towards a general purpose CNN for long range dependencies in ND
DW Romero, DM Knigge, A Gu, EJ Bekkers, E Gavves, JM Tomczak, ...
Workshop on Continuous Time Methods for Machine Learning (ICML 2022), 2022
112022
Modelling Long Range Dependencies in ND: From Task-Specific to a General Purpose CNN
DM Knigge, DW Romero, A Gu, E Gavves, EJ Bekkers, JM Tomczak, ...
International Conference on Learning Representations (ICLR), 2023, 2023
102023
Laughing hyena distillery: Extracting compact recurrences from convolutions
S Massaroli, M Poli, D Fu, H Kumbong, R Parnichkun, D Romero, ...
Advances in Neural Information Processing Systems 36, 2024
22024
Fast, Expressive Equivariant Networks through Weight-Sharing in Position-Orientation Space
EJ Bekkers, S Vadgama, R Hesselink, PA Van der Linden, DW Romero
The Twelfth International Conference on Learning Representations, ICLR 2024, 2023
2*2023
DNArch: Learning Convolutional Neural Architectures by Backpropagation
DW Romero, N Zeghidour
Workshop on Differentiable Almost Everything (ICML), 2023
12023
Self-Supervised Detection of Perfect and Partial Input-Dependent Symmetries
A Urbano, DW Romero
arXiv preprint arXiv:2312.12223, 2023
2023
Learned Gridification for Efficient Point Cloud Processing
PA Van der Linden, DW Romero, EJ Bekkers
arXiv preprint arXiv:2307.14354, 2023
2023
Generative Fourier-Based Auto-encoders: Preliminary Results
A Zonta, A El Hassouni, DW Romero, JM Tomczak
Machine Learning, Optimization, and Data Science: 6th International …, 2020
2020
THE DISTILLERY
D Romero
Sustav trenutno ne može provesti ovu radnju. Pokušajte ponovo kasnije.
Članci 1–18