Methods for neural decoding using machine learning, deep learning, and quantum-inspired algorithms
- Date
- January 17 (Wed) at 15:00 - 16:15, 2024 (JST)
- Speaker
-
- Kei Majima (Researcher, National Institutes for Quantum Science and Technology (QST))
- Venue
- via Zoom
- Language
- English
- Host
- Shinichiro Fujii
Note: The format of this event has changed from hybrid to Zoom only. However, you will still be able to watch it on the screen in Room #359 of the Main Research Building.
(This is a joint seminar with iTHEMS Biology group.)
Recent advances in machine learning have enabled the extraction of intrinsic information from neural activities, a field known as neural decoding. In this presentation, I will introduce several machine learning methods recently developed for neural decoding analysis: 1) a method for visualizing subjective images in the human mind based on brain activity [1], 2) a supervised algorithm designed for predicting discrete ordinal variables [2], and 3) a fast classical algorithm algorithm inspired by quantum computation for approximating principal component analysis (PCA) and canonical correlation analysis (CCA), potentially allowing for the analysis of vast-dimensional neural data [3]. Following these presentations, I am eager to engage in discussions with participants at the RIKEN Quantum Seminar regarding potential collaborations.
References
- N. Koide-Majima, S. Nishimoto, and K. Majima, Mental image reconstruction from human brain activity: Neural decoding of mental imagery via deep neural network-based Bayesian estimation, Neural Networks 170, 349 (2024), doi: 10.1016/j.neunet.2023.11.024
- E. Satake, K. Majima, S. Aoki, and Y. Kamitani, Sparse ordinal logistic regression and its application to brain decoding, Frontiers in Neuroinformatics 12, 51 (2018), doi: 10.3389/fninf.2018.00051
- N. Koide-Majima and K. Majima, Quantum-inspired canonical correlation analysis for exponentially large dimensional data, Neural Networks 135, 55 (2021), doi: 10.1016/j.neunet.2020.11.019
This is a closed event for scientists. Non-scientists are not allowed to attend. If you are not a member or related person and would like to attend, please contact us using the inquiry form. Please note that the event organizer or speaker must authorize your request to attend.