Projects funded by the NCN
Computationally efficient dynamic neural networks
2023/49/N/ST6/02513
Keywords:
deep learning machine learning dynamic neural networks conditional computation computationally efficient deep learning affordable deep learning,
Descriptors:
- ST6_11:
- ST6_07:
Panel:
ST6 - Computer science and informatics: informatics and information systems, computer science, scientific computing, intelligent systems
Host institution :
IDEAS NCBR Sp. z o.o.
woj. mazowieckie
Principal investigator (from the host institution):
Number of co-investigators in the project: 2
Call: PRELUDIUM 22 - announced on 2023-03-16
Amount awarded: 53 070 PLN
Project start date (Y-m-d): 2024-01-24
Project end date (Y-m-d): 2025-01-23
Project duration:: 12 months (the same as in the proposal)
Project status: Project settled
Project description
Download the project description in a pdf file
Note - project descriptions were prepared by the authors of the applications themselves and placed in the system in an unchanged form.
Information in the final report
- Articles in post-conference publications (2)
- Exploiting Activation Sparsity with Dense to Dynamic-k Mixture-of-Experts ConversionAuthors:Filip Szatkowski, Bartosz Wójcik, Mikołaj Piórczyński, Simone ScardapaneConference:Advances in Neural Information Processing Systems 37 (NeurIPS 2024) Main Conference Track (rok: 2025, tom: -, strony: -), Wydawca: -Data:konferencja 10-15.12.2024Status:PublishedDOI:
- Adaptive Computation Modules: Granular Conditional Computation for Efficient InferenceAuthors:Bartosz Wójcik, Alessio Devoto, Karol Pustelnik, Pasquale Minervini, Simone ScardapaneConference:AAAI Technical Track on Machine Learning VI (rok: 2025, tom: The 39th Annual AAAI Conference on Artificial Intelligence, strony: 21510-21518), Wydawca: -Data:konferencja 27.02.2025 - 04.03.2025Status:PublishedDOI:10.1609/aaai.v39i20.35453 - link to the publication