Prof. Dr. Gemma Roig
Department of Computer Science
Computational Vision and AI lab
Office 216, Robert-Mayer-Str. 11-15
60325 Frankfurt am Main
60325 Frankfurt am Main
Germany
+49 (0)69 798-28692
E-Mail: roig@cs.uni-frankfurt.de
Scientific Focus
My research focuses on understanding the underlying computational principles of visual intelligence of humans and artificial systems, with the aim of developing a general artificial intelligence framework. Such general artificial intelligence system, is fundamental to design machine models that mimic or surpass human performance in specific domains, and that can automatically learn new tasks. To this end, I develop computational models using deep learning and state-of-the-art artificial intelligence algorithms and use those to explain behavioural and brain data, including fMRI and EEG.
Methods
Computational modeling, machine learning
Selected Publications
Dwivedi, K., Bonner, M.F., Cichy, R.M., Roig, G. (2021). Unveiling functions of the visual cortex using task-specific deep neural networks. PLoS Comp. Biol., 17 (8), e1009267, https://doi.org/10.1371/journal.pcbi.1009267
Dwivedi, K., Cichy, R.M., Roig, G. (2020). Unravelling Representations in Scene-selective Brain Regions Using Scene Parsing Deep Neural Networks. Journal of Cognitive Neuroscience, 33(10), 2032-2043, https://doi.org/10.1162/jocn_a_01624
Han, Y., Roig, G., Geiger, G., & Poggio, T. (2020). Scale and translation-invariance for novel objects in human vision. Scientific Reports, 10, 1411. doi: 10.1038/s41598-019-57261-6
Dwivedi, K., Roig, G. (2019). Representation Similarity Analysis for Efficient Task taxonomy & Transfer Learning. In Proc. of IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 12379-12388. doi: 10.1109/CVPR.2019.01267
Volokitin, A., Roig, G., & Poggio, T.A. (2017). Do deep neural networks suffer from crowding? Advances in Neural Information Processing Systems, pp. 5628-5638