Sparser, Better, Deeper, Stronger: Improving Static Sparse Training with Exact Orthogonal Initialization.

Aleksandra Nowak 0001 Aleksandra Irena Nowak Jagiellonian University, Faculty of Mathematics and Computer Science, Krakow, Poland https://scholar.google.com/citations?user=2A-eZhQAAAAJ https://orcid.org/0000-0002-2830-6613 Aleksandra Nowak Aleksandra Nowak 0002 Poznan Supercomputing and Networking Center (PSNC), Poznań, Poland https://www.linkedin.com/in/aleksandra-maria-nowak/ https://orcid.org/0009-0005-1274-1783 Aleksandra Nowak 0003 National Research Institute, Central Institute for Labour Protection, Department of Personal Protective Equipment, Lodz, Poland https://orcid.org/0000-0002-4088-327X Aleksandra Nowak 0001 Lukasz Gniecki Filip Szatkowski Jacek Tabor Sparser, Better, Deeper, Stronger: Improving Static Sparse Training with Exact Orthogonal Initialization. 2024 ICML https://proceedings.mlr.press/v235/nowak24a.html https://openreview.net/forum?id=svm53KQAtN conf/icml/2024 db/conf/icml/icml2024.html#0001GST24 38474-38494

Aleksandra Irena Nowak Lukasz Gniecki Filip Szatkowski Jacek Tabor Sparser, Better, Deeper, Stronger: Improving Sparse Training with Exact Orthogonal Initialization. 2024 abs/2406.01755 CoRR https://doi.org/10.48550/arXiv.2406.01755 db/journals/corr/corr2406.html#abs-2406-01755

Aleksandra Irena Nowak Otniel-Bogdan Mercea Anurag Arnab Jonas Pfeiffer Yann N. Dauphin Utku Evci Towards Optimal Adapter Placement for Efficient Transfer Learning. 2024 abs/2410.15858 CoRR https://doi.org/10.48550/arXiv.2410.15858 db/journals/corr/corr2410.html#abs-2410-15858 streams/journals/corr

Aleksandra Nowak 0001 Bram Grooten Decebal Constantin Mocanu Jacek Tabor Fantastic Weights and How to Find Them: Where to Prune in Dynamic Sparse Training. 2023 NeurIPS http://papers.nips.cc/paper_files/paper/2023/hash/ad02c6f3824f871395112ae71a28eff7-Abstract-Conference.html conf/nips/2023 db/conf/nips/neurips2023.html#0001GMT23 Mateusz Olko Michal Zajac 0005 Aleksandra Nowak 0001 Nino Scherrer Yashas Annadani Stefan Bauer Lukasz Kucinski Piotr Milos Trust Your 𝛁: Gradient-based Intervention Targeting for Causal Discovery. 2023 NeurIPS http://papers.nips.cc/paper_files/paper/2023/hash/9eda77f505efbb89462970d739143f73-Abstract-Conference.html conf/nips/2023 db/conf/nips/neurips2023.html#Olko00SABKM23

Aleksandra Nowak 0001 Bram Grooten Decebal Constantin Mocanu Jacek Tabor Fantastic Weights and How to Find Them: Where to Prune in Dynamic Sparse Training. 2023 abs/2306.12230 CoRR https://doi.org/10.48550/arXiv.2306.12230 db/journals/corr/corr2306.html#abs-2306-12230

Andrzej Bedychaj Przemyslaw Spurek Aleksandra Nowak 0001 Jacek Tabor Nonlinear Weighted Independent Component Analysis. 3-16 2022 IPMU (2) https://doi.org/10.1007/978-3-031-08974-9_1 conf/ipmu/2022-2 db/conf/ipmu/ipmu2022-2.html#BedychajS0T22 Lukasz Maziarka Aleksandra Nowak 0001 Maciej Wolczyk Andrzej Bedychaj On the Relationship Between Disentanglement and Multi-task Learning. 625-641 2022 ECML/PKDD (1) https://doi.org/10.1007/978-3-031-26387-3_38 conf/pkdd/2022-1 db/conf/pkdd/pkdd2022-1.html#MaziarkaNWB22 Aleksandra Nowak 0001 Romuald A. Janik Discovering Wiring Patterns Influencing Neural Network Performance. 631-646 2022 ECML/PKDD (3) https://doi.org/10.1007/978-3-031-26409-2_38 conf/pkdd/2022-3 db/conf/pkdd/pkdd2022-3.html#NowakJ22

Mateusz Olko Michal Zajac 0005 Aleksandra Nowak 0001 Nino Scherrer Yashas Annadani Stefan Bauer Lukasz Kucinski Piotr Milos Trust Your ∇: Gradient-based Intervention Targeting for Causal Discovery. 2022 abs/2211.13715 CoRR https://doi.org/10.48550/arXiv.2211.13715 db/journals/corr/corr2211.html#abs-2211-13715

Marcin Sendera Jacek Tabor Aleksandra Nowak 0001 Andrzej Bedychaj Massimiliano Patacchiola Tomasz Trzcinski Przemyslaw Spurek Maciej Zieba Non-Gaussian Gaussian Processes for Few-Shot Regression. 10285-10298 2021 NeurIPS https://proceedings.neurips.cc/paper/2021/hash/54f3bc04830d762a3b56a789b6ff62df-Abstract.html conf/nips/2021 db/conf/nips/neurips2021.html#SenderaTNBPTSZ21

Lukasz Maziarka Aleksandra Nowak 0001 Maciej Wolczyk Andrzej Bedychaj On the relationship between disentanglement and multi-task learning. 2021 abs/2110.03498 CoRR https://arxiv.org/abs/2110.03498 db/journals/corr/corr2110.html#abs-2110-03498

Marcin Sendera Jacek Tabor Aleksandra Nowak 0001 Andrzej Bedychaj Massimiliano Patacchiola Tomasz Trzcinski Przemyslaw Spurek Maciej Zieba Non-Gaussian Gaussian Processes for Few-Shot Regression. 2021 abs/2110.13561 CoRR https://arxiv.org/abs/2110.13561 db/journals/corr/corr2110.html#abs-2110-13561

Przemyslaw Spurek Aleksandra Nowak 0001 Jacek Tabor Lukasz Maziarka Stanislaw Jastrzebski Non-linear ICA Based on Cramer-Wold Metric. 294-305 2020 ICONIP (3) https://doi.org/10.1007/978-3-030-63836-8_25 https://www.wikidata.org/entity/Q112225887 conf/iconip/2020-3 db/conf/iconip/iconip2020-3.html#SpurekNTMJ20

Andrzej Bedychaj Przemyslaw Spurek Aleksandra Nowak 0001 Jacek Tabor WICA: nonlinear weighted ICA. 2020 abs/2001.04147 CoRR https://arxiv.org/abs/2001.04147 db/journals/corr/corr2001.html#abs-2001-04147

Romuald A. Janik Aleksandra Nowak 0001 Neural Networks on Random Graphs. 2020 abs/2002.08104 CoRR https://arxiv.org/abs/2002.08104 db/journals/corr/corr2002.html#abs-2002-08104

Romuald A. Janik Aleksandra Nowak 0001 Neural networks adapting to datasets: learning network size and topology. 2020 abs/2006.12195 CoRR https://arxiv.org/abs/2006.12195 db/journals/corr/corr2006.html#abs-2006-12195

Lukasz Maziarka Marek Smieja Aleksandra Nowak 0001 Jacek Tabor Lukasz Struski Przemyslaw Spurek Set Aggregation Network as a Trainable Pooling Layer. 419-431 2019 ICONIP (2) https://doi.org/10.1007/978-3-030-36711-4_35 conf/iconip/2019-2 db/conf/iconip/iconip2019-2.html#MaziarkaSNTSS19

Przemyslaw Spurek Aleksandra Nowak 0001 Jacek Tabor Lukasz Maziarka Stanislaw Jastrzebski Non-linear ICA based on Cramer-Wold metric. 2019 abs/1903.00201 CoRR http://arxiv.org/abs/1903.00201 db/journals/corr/corr1903.html#abs-1903-00201

Lukasz Struski Jacek Tabor Igor T. Podolak Aleksandra Nowak 0001 Interpolation in generative models. 2019 abs/1904.03445 CoRR http://arxiv.org/abs/1904.03445 db/journals/corr/corr1904.html#abs-1904-03445

Lukasz Maziarka Marek Smieja Aleksandra Nowak 0001 Jacek Tabor Lukasz Struski Przemyslaw Spurek Deep processing of structured data. 2018 abs/1810.01868 CoRR http://arxiv.org/abs/1810.01868 db/journals/corr/corr1810.html#abs-1810-01868

Yashas Annadani Anurag Arnab Stefan Bauer Andrzej Bedychaj Yann N. Dauphin Utku Evci Lukasz Gniecki Bram Grooten Romuald A. Janik Stanislaw Jastrzebski Lukasz Kucinski Lukasz Maziarka Otniel-Bogdan Mercea Piotr Milos Decebal Constantin Mocanu Mateusz Olko Massimiliano Patacchiola Jonas Pfeiffer Igor T. Podolak Nino Scherrer Marcin Sendera Marek Smieja Przemyslaw Spurek Lukasz Struski Filip Szatkowski Jacek Tabor Tomasz Trzcinski Maciej Wolczyk Michal Zajac 0005 Maciej Zieba