Følg
Christos Thrampoulidis
Christos Thrampoulidis
Assistant Professor, University of British Columbia, ECE Department
Verificeret mail på ucsb.edu - Startside
Titel
Citeret af
Citeret af
År
Precise Error Analysis of Regularized -Estimators in High Dimensions
C Thrampoulidis, E Abbasi, B Hassibi
IEEE Transactions on Information Theory 64 (8), 5592-5628, 2018
3022018
Regularized linear regression: A precise analysis of the estimation error
C Thrampoulidis, S Oymak, B Hassibi
Conference on Learning Theory, 1683-1709, 2015
257*2015
A model of double descent for high-dimensional binary linear classification
Z Deng, A Kammoun, C Thrampoulidis
Information and Inference: A Journal of the IMA 11 (2), 435-495, 2022
2062022
The squared-error of generalized lasso: A precise analysis
S Oymak, C Thrampoulidis, B Hassibi
2013 51st Annual Allerton Conference on Communication, Control, and …, 2013
1552013
Linear stochastic bandits under safety constraints
S Amani, M Alizadeh, C Thrampoulidis
Advances in Neural Information Processing Systems, 9256-9266, 2019
1362019
Lasso with non-linear measurements is equivalent to one with linear measurements
C Thrampoulidis, E Abbasi, B Hassibi
Advances in Neural Information Processing Systems 28, 2015
1302015
Optimal placement of distributed energy storage in power networks
C Thrampoulidis, S Bose, B Hassibi
IEEE Transactions on Automatic Control 61 (2), 416-429, 2015
1292015
Label-imbalanced and group-sensitive classification under overparameterization
GR Kini, O Paraskevas, S Oymak, C Thrampoulidis
Advances in Neural Information Processing Systems 34, 18970-18983, 2021
1022021
Revealing hidden scenes by photon-efficient occlusion-based opportunistic active imaging
F Xu, G Shulkind, C Thrampoulidis, JH Shapiro, A Torralba, FNC Wong, ...
Optics express 26 (8), 9945-9962, 2018
932018
Exploiting occlusion in non-line-of-sight active imaging
C Thrampoulidis, G Shulkind, F Xu, WT Freeman, JH Shapiro, A Torralba, ...
IEEE Transactions on Computational Imaging 4 (3), 419-431, 2018
842018
Transformers as support vector machines
DA Tarzanagh, Y Li, C Thrampoulidis, S Oymak
arXiv preprint arXiv:2308.16898, 2023
802023
Using unknown occluders to recover hidden scenes
AB Yedidia, M Baradad, C Thrampoulidis, WT Freeman, GW Wornell
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern …, 2019
792019
Fednest: Federated bilevel, minimax, and compositional optimization
DA Tarzanagh, M Li, C Thrampoulidis, S Oymak
International Conference on Machine Learning, 21146-21179, 2022
772022
Analytic study of double descent in binary classification: The impact of loss
GR Kini, C Thrampoulidis
2020 IEEE International Symposium on Information Theory (ISIT), 2527-2532, 2020
762020
Autobalance: Optimized loss functions for imbalanced data
M Li, X Zhang, C Thrampoulidis, J Chen, S Oymak
Advances in Neural Information Processing Systems 34, 3163-3177, 2021
742021
Imbalance trouble: Revisiting neural-collapse geometry
C Thrampoulidis, GR Kini, V Vakilian, T Behnia
Advances in Neural Information Processing Systems 35, 27225-27238, 2022
722022
Symbol error rate performance of box-relaxation decoders in massive mimo
C Thrampoulidis, W Xu, B Hassibi
IEEE Transactions on Signal Processing 66 (13), 3377-3392, 2018
722018
Provable benefits of overparameterization in model compression: From double descent to pruning neural networks
X Chang, Y Li, S Oymak, C Thrampoulidis
Proceedings of the AAAI Conference on Artificial Intelligence 35 (8), 6974-6983, 2021
622021
Benign overfitting in multiclass classification: All roads lead to interpolation
K Wang, V Muthukumar, C Thrampoulidis
Advances in Neural Information Processing Systems 34, 24164-24179, 2021
612021
Improved bounds on Gaussian MAC and sparse regression via Gaussian inequalities
I Zadik, Y Polyanskiy, C Thrampoulidis
2019 IEEE International Symposium on Information Theory (ISIT), 430-434, 2019
562019
Systemet kan ikke foretage handlingen nu. Prøv igen senere.
Artikler 1–20