Lecture + Tutorial, Summer 2024Computational Neuroscience:
|
Lecture:
Richard Kempter,
Benjamin Blankertz
Tutorial on analysis of spike data:
Gaspar Cano, Stefano Masserini, Paula Kuokkanen; grading: Ikhwan Bin Khalid
Tutorial on analysis of EEG data:
Benjamin Blankertz and coworkers
Date: From 18-April-2024 to 18-July-2024.
Location: Lectures and tutorials take place at the Bernstein Center for Computational Neuroscience Berlin.
Times:
Lectures (2 SWS, 2 ECTS): Thursdays from 09:15 am to 10:45 am,
Lecture Hall 102, Haus 6
Tutorials (2 SWS, 5 ECTS): Thursdays from 11:00 am to 12:30 pm,
ITB seminar room 012, Haus 4/ Computer Pool, Haus 2
Target Group: Students of Computational Neuroscience, Medical Neuroscience, Biology, Biophysics, Physics, Mathematics, and Computer Science.
Requirements: Basic knowledge in Neurobiology and Mathematics at the level of the first semester of the Masters Program in Computational Neuroscience.
Topics: This course is the second semester of the module
''Acquisition and Analysis of Neural Data'' of the
Master Program in Computational Neuroscience, and this second part
focuses on statistical analyses
of neural data:
(1) Analysis of spike trains (spike statistics, neural coding, theory of point processes, linear systems theory, correlation analysis, spike-triggered average, reverse correlation, STRF, neural decoding, signal-detection theory, information theory, signal-to-noise ratio analysis). This Moodle Course provides details on the tutorials.
(2) Statistical analysis of electroencephalogram (EEG) data, e.g., forward calculation; investigation of event-related potentials (ERPs) and event-related desynchronization (ERD); spatial filters; classification.
(01) | 18.04. | The firing rate (Richard Kempter, RK) |
(02) | 25.04. | Spike-train statistics: spike-triggered average and reverse correlation (RK, seminar room in Haus 4!) |
(03) | 02.05. | Correlation functions and the neural code (RK) |
09.05. | - Himmelfahrt, no lecture - | |
(04) | 16.05. | Neural encoding (RK) |
(05) | 23.05. | Neural decoding (RK) |
(06) | 30.05. | Information theory (RK) |
(07) | 06.06. | Overview BCI; Characterization of Gaussian Distributions (Benjamin Blankertz, BB) |
(08) | 13.06. | ERP-based BCIs; Spatio-Temporal Features; LDA (BB) |
(09) | 20.06. | Shrinkage of the Empirical Covariance Matrix (BB) |
(10) | 27.06. | Linear Model of EEG; Spatial Patterns and Spatial Filters (BB) |
(11) | 04.07. | Modulations of Brain Rhythms; Spectral Features (BB) |
(12) | 11.07. | Common Spatial Patterns Analysis (BB) |
(13) | 18.07. | Special BCI Topics and Wrap-Up (BB) |
Course Certificates: To obtain a course certificate, at least 60% of the points in the weekly exercises must be obtained. This 60% rule applies to each of the two blocks in the summer term separately. Thus, 60% must be achieved on average over all assignments in the part on the "Analysis of spike trains" and, separately, in the part on the "Statistical analysis of EEG". Participants via the Master module MB-B12 (HU Berlin, Biology) need to obtain only 50%.
In the part on the "Analysis of spike trains", students must work in groups of 2 or 3 and submit a single joint solution for each problem set. Group composition should change once (groups of size 3 do not keep subgroups of 2 students). Groups change ideally inbetween the two blocks of problem sets 1-3 and 4-6. An active participation in the tutorials is strongly encouraged. The successful presentation of a problem will be rewarded by bonus points (10%) for the presenting student. Furthermore, a successful participation requires that one (out of two) very short written assessments (typically et the end of the blocks) is passed.
The specific rules on the part on the "Statistical Analysis of EEG" will be communicated separately.
Exam: The final oral exam on the module "Acquisition and Analysis of Neuronal Data" will take place on September 12/13, 2024.
Background material for the analysis of spike trains:
P. Dayan and L.F. Abbott (2001)
Theoretical Neuroscience.
MIT Press, Cambridge, Massachusetts.