Ghada Sokar, MSc

Onderzoeksoutput

Onderzoeksoutput per jaar 2019 2019 2020 2021 2022 2023 2024

• 11 Conferentiebijdrage
• 3 Tijdschriftartikel
• 2 Paper
• 1 Dissertatie 1 (Onderzoek TU/e / Promotie TU/e)

Onderzoeksoutput per jaar

Onderzoeksoutput per jaar

Zoekresultaten

• 2024

  Continual Learning with Dynamic Sparse Training: Exploring Algorithms for Effective Model Updates

  Yildirim, M. O., Gok Yildirim, E. C., Sokar, G., Mocanu, D. C. & Vanschoren, J., 2024, 1st Conference on Parsimony and Learning, CPAL 2024Hongkong3 January 2024through 6 January 2024. PMLR, blz. 94-107 14 blz. (Proceedings of Machine Learning Research; vol. 234).

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

  Open Access

  • Sparsity 100%
  • Optimal Configuration 33%
  • Neural Network 33%
  • Regularization 33%
  1 Citaat (Scopus)
• 2023

  Automatic Noise Filtering with Dynamic Sparse Training in Deep Reinforcement Learning

  Grooten, B., Sokar, G., Dohare, S., Mocanu, E., Taylor, M. E., Pechenizkiy, M. & Mocanu, D. C., 2023, AAMAS '23: Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems. Richland: International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS), blz. 1932-1941 10 blz.

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

  • Reinforcement Learning 100%
  • Deep Reinforcement Learning 100%
  • Learning Algorithm 100%
  • Learning Transfer 100%
  • Robot 66%
  4 Citaten (Scopus)

  1 Downloads (Pure)

• Avoiding Forgetting and Allowing Forward Transfer in Continual Learning via Sparse Networks.

  Sokar, G., Mocanu, D. C. & Pechenizkiy, M., 2023, Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2022, Proceedings. Amini, M.-R., Canu, S., Fischer, A., Guns, T., Kralj Novak, P. & Tsoumakas, G. (uitgave). blz. 85-101 17 blz. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); vol. 13715 LNAI).

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

  • Subnetwork 100%
  • Capacity Model 100%
  • Neural Network 50%
  • Desirable Property 50%

• Learning Continually Under Changing Data Distributions

  Sokar, G., 31 okt. 2023, Eindhoven: Eindhoven University of Technology. 155 blz.

  Onderzoeksoutput: Scriptie › Dissertatie 1 (Onderzoek TU/e / Promotie TU/e)

  Open Access

  Bestand
  344 Downloads (Pure)
• 2022

  Deep Ensembling with No Overhead for either Training or Testing: The All-Round Blessings of Dynamic Sparsity

  Liu, S., Chen, T., Atashgahi, Z., Chen, X., Sokar, G., Mocanu, E., Pechenizkiy, M., Wang, Z. & Mocanu, D. C., 20 jan. 2022, (Geaccepteerd/In druk) International Conference on Learning Representations, ICLR 2022.

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

  Bestand

  • Sparsity 100%
  • Subnetwork 100%
  • Learning System 100%
  • Deep Neural Network 100%
  • Neural Network 100%
  104 Downloads (Pure)

• Dynamic Sparse Training for Deep Reinforcement Learning

  Sokar, G., Mocanu, E., Mocanu, D. C., Pechenizkiy, M. & Stone, P., 2022.

  Onderzoeksoutput: Bijdrage aan congres › Paper › Academic

  • Deep Reinforcement Learning 100%
  • Reinforcement Learning 40%
  • Learning Agent 40%
  • Neural Network 40%
  • Learning Process 20%

• Quick and robust feature selection: the strength of energy-efficient sparse training for autoencoders

  Atashgahi, Z. (Corresponding author), Sokar, G., van der Lee, T., Mocanu, E., Mocanu, D. C., Veldhuis, R. & Pechenizkiy, M., jan. 2022, In: Machine Learning. 111, 1, blz. 377-414 38 blz.

  Onderzoeksoutput: Bijdrage aan tijdschrift › Tijdschriftartikel › Academic › peer review

  Open Access

  Bestand

  • Energy Efficient 100%
  • Feature Selection 100%
  • Autoencoder 100%
  • Sparsity 16%
  • High Dimensional Data 16%
  9 Citaten (Scopus)

  144 Downloads (Pure)

• Where to pay attention in sparse training for feature selection?

  Sokar, G., Atashgahi, Z., Pechenizkiy, M. & Mocanu, D. C., 2022, 36th Conference on Neural Information Processing Systems (NeurIPS 2022).

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review
• 2021

  Continual Lifelong Learning for Intelligent Agents

  Sokar, G., 2021, International Joint Conference on Artificial Intelligence (IJCAI-21).

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

• FreeTickets: Accurate, Robust and Efficient Deep Ensemble by Training with Dynamic Sparsity

  Liu, S., Chen, T., Atashgahi, Z., Chen, X., Sokar, G., Mocanu, E., Pechenizkiy, M., Wang, Z. & Mocanu, D. C., 17 mrt. 2021, (Ingediend).

  Onderzoeksoutput: Bijdrage aan congres › Paper › Academic

  • Sparsity 100%
  • Neural Network 100%
  • Financial Concern 25%
  • Ensemble Method 25%
  • Uncertainty Estimation 25%

• SpaceNet: Make Free Space For Continual Learning

  Sokar, G. (Corresponding author), Mocanu, D. C. & Pechenizkiy, M., 7 jun. 2021, In: Neurocomputing. 439, blz. 1-11 11 blz.

  Onderzoeksoutput: Bijdrage aan tijdschrift › Tijdschriftartikel › Academic › peer review

  Open Access

  Bestand

  • Neural Network 100%
  • Class-Incremental Learning 100%
  • Capacity Model 50%
  36 Citaten (Scopus)

  116 Downloads (Pure)

• Topological Insights into Sparse Neural Networks

  Liu, S., van der Lee, T. T., Yaman, A., Atashgahi, Z., Ferrar, D., Sokar, G., Pechenizkiy, M. & Mocanu, D. C., 2021, Machine Learning and Knowledge Discovery in Databases : European Conference, ECML PKDD 2020, Ghent, Belgium, September 14–18, 2020, Proceedings, Part III. Hutter, F., Kersting, K., Lijffijt, J. & Valera, I. (uitgave). Cham: Springer, blz. 279-294 16 blz. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); vol. 12459 LNAI).

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

  Open Access

  Bestand

  • Neural Network 100%
  • Graph Theory 25%
  • Deep Neural Network 12%
  • Network Topology 12%
  • Resource Requirement 12%
  5 Citaten (Scopus)

  140 Downloads (Pure)
• 2020

  Learning Invariant Representation for Continual Learning

  Sokar, G., Mocanu, D. C. & Pechenizkiy, M., 21 dec. 2020, (Geaccepteerd/In druk) AAAI-21 International Workshop on Meta-Learning for Computer Vision (MeL4CV), 2021.

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

  Bestand

  • Invariant Representation 100%
  • Knowledge Transfer 20%
  • Intelligent Agent 20%
  • Class Invariant 20%
  • Experimental Result 20%
  210 Downloads (Pure)

• Quick and Robust Feature Selection: the Strength of Energy-efficient Sparse Training for Autoencoders

  Atashgahi, Z. (Corresponding author), Sokar, G., van der Lee, T., Mocanu, E., Mocanu, D. C., Veldhuis, R. & Pechenizkiy, M., 1 dec. 2020, In: arXiv. 2020, 29 blz., 2012.00560.

  Onderzoeksoutput: Bijdrage aan tijdschrift › Tijdschriftartikel › Academic

  Open Access

  Bestand

  • Energy Efficient 100%
  • Feature Selection 100%
  • Autoencoder 100%
  • Sparsity 16%
  • High Dimensional Data 16%
  133 Downloads (Pure)

• Self-Attention Meta-Learner for Continual Learning

  Sokar, G., Mocanu, D. C. & Pechenizkiy, M., 17 dec. 2020, (Geaccepteerd/In druk) 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021).

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

  Bestand

  • Attention (Machine Learning) 100%
  • Neural Network 100%
  • Intelligent Agent 50%
  • Multiple Task 50%
  • Starting Point 50%
  87 Downloads (Pure)
• 2019

  A Generic OCR Using Deep Siamese Convolution Neural Networks

  Sokar, G., Hemayed, E. E. & Rehan, M., 16 jan. 2019, 2018 IEEE 9th Annual Information Technology, Electronics and Mobile Communication Conference, IEMCON 2018. Chakrabarti, S. & Saha, H. N. (uitgave). Institute of Electrical and Electronics Engineers, blz. 1238-1244 7 blz. 8614784

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

  • Convolutional Neural Network 100%
  • Optical Character Recognition 100%
  • Recognition System 66%
  • Classification Task 50%
  • Deep Neural Network 50%
  17 Citaten (Scopus)

• IP Session on Machine Learning Applications in IC Test-Related Tasks

  Sokar, G., Zakaria, Y., Rabie, A., Madkour, K., Leventhal, I., Rivoir, J., Gu, X. & Stratigopoulos, H. G., apr. 2019, 2019 IEEE 37th VLSI Test Symposium, VTS 2019. IEEE Computer Society, 1 blz. 8758634

  Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

  • Learning System 100%
  • Deep Learning 25%