mgr Tomasz Danel

Jednostki:

  • Wydział Matematyki i Informatyki UJ
  • Instytut Informatyki i Matematyki Komputerowej
  • Katedra Uczenia Maszynowego

Publikacje:

8.
HuggingMolecules: an open-source library for transformer-based molecular property prediction, INTERNATIONAL CONFERENCE ON LEARNING REPRESENTATIONS (WORKSHOP TRACK) (2021),
7.
Multitask Learning Using BERT with Task-Embedded Attention, IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS [IJCNN], (2021),
6.
Comparison of Atom Representations in Graph Neural Networks for Molecular Property Prediction, IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS [IJCNN], (2021),
5.
Processing of Incomplete Images by (Graph) Convolutional Neural Networks, INTERNATIONAL CONFERENCE ON NEURAL INFORMATION PROCESSING vol. Lecture Notes in Computer Science, volume 12533 (2020), 512-523
4.
Spatial Graph Convolutional Networks, INTERNATIONAL CONFERENCE ON NEURAL INFORMATION PROCESSING vol. Communications in Computer and Information Science book series (CCIS, volume 1333) (2020), 668-675
1.
Tomasz Danel, Stanisław Jastrzębski, Łukasz Maziarka, Sławomir Mucha, Krzysztof Rataj, Jacek Tabor
Molecule-Augmented Attention Transformer, NEURAL INFORMATION PROCESSING SYSTEMS (WORKSHOP TRACK) (2019),

Konferencje:

3.
34th Conference on Neural Information Processing Systems [NeurIPS 2020], Neural Information Processing Systems Foundation, [virtual], [virtual], 2020-12-06 - 2020-12-12
2.
ML in PL Conference, Uniwersytet Warszawski, Warszawa, Polska, 2019-11-22 - 2019-11-25

Granty (realizowane po maju 2009 roku)

TytułRolaRozpoczęcieZakończenie
Połączenie symulacji molekularnej i uczenia głębokiego w projektowaniu leków de novoKierownik2021-02-012023-01-31