Rekurentní hluboké neuronové sítě s biologicky realistickou architekturou
| Thesis title in Czech: | Rekurentní hluboké neuronové sítě s biologicky realistickou architekturou |
|---|---|
| Thesis title in English: | Recurrent deep neural networks with biologically realistic architecture |
| Academic year of topic announcement: | 2023/2024 |
| Thesis type: | diploma thesis |
| Thesis language: | |
| Department: | Department of Software and Computer Science Education (32-KSVI) |
| Supervisor: | Mgr. Ján Antolík, Ph.D. |
| Author: | hidden - assigned and confirmed by the Study Dept. |
| Date of registration: | 04.12.2023 |
| Date of assignment: | 05.12.2023 |
| Confirmed by Study dept. on: | 05.12.2023 |
| Guidelines |
| Application of deep neural networks to large datasets of neural data recorded in response to library of visual stimuly become the dominant method of unraveling the function of neurons in visual cortex. The standard approaches however (i) ignore known anatomical structure of visual cortex, (ii) use purely feed-forward NN as opposed to the intrinsically recurrent biological networks, (iii) only capture the mean steady state response. To address this, in this project you will build a DNN model composed of multiple recurrent neural network stages, that will be constrained to follow various know anatomical structures, and train the resulting model on fine temporal recordings of V1 responses to natural images. |
| References |
| [1] Antolík, J, Monier, C., Davison, A., Frégnac Y.; A comprehensive data-driven model of cat primary visual cortex. bioRxiv, 416156
[2] Antolík, J., & Davison, A. P. (2013). Integrated workflows for spiking neuronal network simulations. Frontiers in Neuroinformatics, 7(December), 1–15. https://doi.org/10.3389/fninf.2013.00034 [1] E. Margalit, H. Lee, D. Finzi, J. J. DiCarlo, K. Grill-Spector, and D. L. K. Yamins, A Unifying Principle for the Functional Organization of Visual Cortex, bioRxiv, preprint, May 2023. doi: 10.1101/2023.05.18.541361. |