001041677 001__ 1041677
001041677 005__ 20250505202225.0
001041677 037__ $$aFZJ-2025-02382
001041677 041__ $$aEnglish
001041677 1001_ $$0P:(DE-Juel1)178725$$aBouss, Peter$$b0$$eCorresponding author$$ufzj
001041677 1112_ $$aComputational Neuroscience Academy$$cKrakow$$d2025-07-17 - 2025-07-23$$gCNA 2023$$wPoland
001041677 245__ $$aNormalizing flows for nonlinear dimensionality reduction ofelectrophysiological recordings
001041677 260__ $$c2023
001041677 3367_ $$033$$2EndNote$$aConference Paper
001041677 3367_ $$2BibTeX$$aINPROCEEDINGS
001041677 3367_ $$2DRIVER$$aconferenceObject
001041677 3367_ $$2ORCID$$aCONFERENCE_POSTER
001041677 3367_ $$2DataCite$$aOutput Types/Conference Poster
001041677 3367_ $$0PUB:(DE-HGF)24$$2PUB:(DE-HGF)$$aPoster$$bposter$$mposter$$s1746445560_28061$$xAfter Call
001041677 520__ $$aEven though the cortex has many active neurons, neuronal populations for different brain areasshould dwell on a low-dimensional manifold [1]. Principal component analysis versions are used toestimate this manifold and its dimension. Although successful, these methods assume that the data iswell described by a Gaussian distribution and ignore features like skewness and bimodality. Therefore,they perform poorly as generative models.Normalizing Flows (NFs) allow us to learn neural activity statistics and generate artificial samples [2,3]. These neural networks learn a dimension-preserving estimator of the data’s probability distribution.They are simpler than generative adversarial networks (GANs) and variational autoencoders (VAEs)since they learn only one bijective mapping and can compute the likelihood correctly due to tractableJacobians at each building block.NFs are trained to distinguish relevant (in manifold) from noisy dimensions (out of manifold). To dothis, we break the original symmetry of the latent space by pushing maximal variance of the data to becaptured by as few dimensions as possible — the same idea underpinning PCA, a linear model, adoptedhere for nonlinear mappings. NFs’ unique characteristics allows us to estimate the neural manifold’sdimensions and describe the underlying manifold without discarding any information.Our adaptation is validated on simulated datasets of various complexity created using a hidden man-ifold model with specified dimensions. Reconstructing data with a few latent NF dimensions shows ourapproach’s capability. In this case, our nonlinear approaches outperform linear ones. We identify mani-folds in high-gamma EEG recordings using the aforementioned technique. In the experiment of [4], 128electrodes recorded during four movement tasks. These data show a heavy-tailed distribution along someof the first principal components. NFs can learn higher-order correlations while linear models like PCAare limited to Gaussian statistics. We can also better match features to latent dimensions by flatteningthe latent space. We now have fewer latent dimensions that explain most data variance.References[1] J. A. Gallego, M. G. Perich, L. E. Miller, and S. A. Solla, Neuron, 94(5), 978-984 (2017).[2] L. Dinh, D. Krueger, and Y. Bengio, In Conference on Learning Representations, ICLR (2015).[3] L. Dinh, J. Sohl-Dickstein, and S. Bengio, In Conference on Learning Representations, ICLR (2017).[4] R. T. Schirrmeister, J. T. Springenberg, L. D. J. Fiederer, M. Glasstetter, K. Eggensperger, M. Tanger-mann, ... and T. Ball, Human brain mapping, 38(11), 5391-5420 (2017).
001041677 536__ $$0G:(DE-HGF)POF4-5232$$a5232 - Computational Principles (POF4-523)$$cPOF4-523$$fPOF IV$$x0
001041677 536__ $$0G:(DE-HGF)POF4-5234$$a5234 - Emerging NC Architectures (POF4-523)$$cPOF4-523$$fPOF IV$$x1
001041677 536__ $$0G:(GEPRIS)368482240$$aGRK 2416 - GRK 2416: MultiSenses-MultiScales: Neue Ansätze zur Aufklärung neuronaler multisensorischer Integration (368482240)$$c368482240$$x2
001041677 536__ $$0G:(DE-Juel-1)BMBF-01IS19077A$$aRenormalizedFlows - Transparent Deep Learning with Renormalized Flows (BMBF-01IS19077A)$$cBMBF-01IS19077A$$x3
001041677 7001_ $$0P:(DE-Juel1)204281$$aNestler, Sandra$$b1
001041677 7001_ $$0P:(DE-Juel1)180150$$aFischer, Kirsten$$b2$$ufzj
001041677 7001_ $$0P:(DE-Juel1)184900$$aMerger, Claudia Lioba$$b3
001041677 7001_ $$0P:(DE-Juel1)178936$$aRene, Alexandre$$b4
001041677 7001_ $$0P:(DE-Juel1)144806$$aHelias, Moritz$$b5$$ufzj
001041677 8564_ $$uhttps://cna2023.ift.uj.edu.pl/documents/152426817/153896610/CNA_2023_BOA_Posters.pdf/c2ab4ff9-fd6e-4290-81bd-1b2dcac9d189
001041677 909CO $$ooai:juser.fz-juelich.de:1041677$$pVDB
001041677 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)178725$$aForschungszentrum Jülich$$b0$$kFZJ
001041677 9101_ $$0I:(DE-588b)36225-6$$6P:(DE-Juel1)178725$$aRWTH Aachen$$b0$$kRWTH
001041677 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)204281$$aForschungszentrum Jülich$$b1$$kFZJ
001041677 9101_ $$0I:(DE-588b)36225-6$$6P:(DE-Juel1)204281$$aRWTH Aachen$$b1$$kRWTH
001041677 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)180150$$aForschungszentrum Jülich$$b2$$kFZJ
001041677 9101_ $$0I:(DE-588b)36225-6$$6P:(DE-Juel1)180150$$aRWTH Aachen$$b2$$kRWTH
001041677 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)184900$$aForschungszentrum Jülich$$b3$$kFZJ
001041677 9101_ $$0I:(DE-588b)36225-6$$6P:(DE-Juel1)184900$$aRWTH Aachen$$b3$$kRWTH
001041677 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)178936$$aForschungszentrum Jülich$$b4$$kFZJ
001041677 9101_ $$0I:(DE-588b)36225-6$$6P:(DE-Juel1)178936$$aRWTH Aachen$$b4$$kRWTH
001041677 9101_ $$0I:(DE-HGF)0$$6P:(DE-Juel1)178936$$a University of Ottawa, Canada$$b4
001041677 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)144806$$aForschungszentrum Jülich$$b5$$kFZJ
001041677 9101_ $$0I:(DE-588b)36225-6$$6P:(DE-Juel1)144806$$aRWTH Aachen$$b5$$kRWTH
001041677 9131_ $$0G:(DE-HGF)POF4-523$$1G:(DE-HGF)POF4-520$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5232$$aDE-HGF$$bKey Technologies$$lNatural, Artificial and Cognitive Information Processing$$vNeuromorphic Computing and Network Dynamics$$x0
001041677 9131_ $$0G:(DE-HGF)POF4-523$$1G:(DE-HGF)POF4-520$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5234$$aDE-HGF$$bKey Technologies$$lNatural, Artificial and Cognitive Information Processing$$vNeuromorphic Computing and Network Dynamics$$x1
001041677 920__ $$lyes
001041677 9201_ $$0I:(DE-Juel1)INM-6-20090406$$kINM-6$$lComputational and Systems Neuroscience$$x0
001041677 9201_ $$0I:(DE-Juel1)IAS-6-20130828$$kIAS-6$$lComputational and Systems Neuroscience$$x1
001041677 980__ $$aposter
001041677 980__ $$aVDB
001041677 980__ $$aI:(DE-Juel1)INM-6-20090406
001041677 980__ $$aI:(DE-Juel1)IAS-6-20130828
001041677 980__ $$aUNRESTRICTED