Dual missions of neural dimensionality reduction
Event details
| Date | 02.06.2023 |
| Hour | 14:00 › 16:00 |
| Speaker | Shuqi Wang |
| Location | |
| Category | Conferences - Seminars |
EDIC candidacy exam
Exam president: Prof. Michael Gastpar
Thesis advisor: Prof. Wulfram Gerstner
Co-examiner: Prof. Mackenzie Mathis
Abstract
Over the last decades, as more neurons are being simultaneously recorded, researchers embrace the new opportunity of analyzing how neural circuits coordinate as a whole to drive behavior. Well suited to this mission, various dimensionality reduction methods have been developed and many have demonstrated the ability to find latent variables that correlate with behavior. In this report, I will review three representative methods and highlight their key methodologies. Furthermore, in addition to building the link to behavior, I will argue that there is another equally important but far less explored mission of neural dimensionality reduction, which is to provide insights into the connectivity structure.
Background papers
Exam president: Prof. Michael Gastpar
Thesis advisor: Prof. Wulfram Gerstner
Co-examiner: Prof. Mackenzie Mathis
Abstract
Over the last decades, as more neurons are being simultaneously recorded, researchers embrace the new opportunity of analyzing how neural circuits coordinate as a whole to drive behavior. Well suited to this mission, various dimensionality reduction methods have been developed and many have demonstrated the ability to find latent variables that correlate with behavior. In this report, I will review three representative methods and highlight their key methodologies. Furthermore, in addition to building the link to behavior, I will argue that there is another equally important but far less explored mission of neural dimensionality reduction, which is to provide insights into the connectivity structure.
Background papers
- Pandarinath, Chethan, et al. "Inferring single-trial neural population dynamics using sequential auto-encoders." Nature methods 15.10 (2018): 805-815. https://www.nature.com/articles/s41592-018-0109-9
- Schneider, Steffen, Jin Hwa Lee, and Mackenzie Weygandt Mathis. "Learnable latent embeddings for joint behavioral and neural analysis." arXiv preprint arXiv:2204.00673 (2022). https://arxiv.org/abs/2204.00673
- Yu, Byron M., et al. "Gaussian-process factor analysis for low-dimensional single-trial analysis of neural population activity." Advances in neural information processing systems 21 (2008). https://proceedings.neurips.cc/paper/2008/hash/ad972f10e0800b49d76fed33a21f6698-Abstract.html
Practical information
- General public
- Free