{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T04:46:15Z","timestamp":1761108375042,"version":"build-2065373602"},"reference-count":57,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2022,1,19]],"date-time":"2022-01-19T00:00:00Z","timestamp":1642550400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61836003","61573150","61573152"],"award-info":[{"award-number":["61836003","61573150","61573152"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>Latent variable models (LVMs) for neural population spikes have revealed informative low-dimensional dynamics about the neural data and have become powerful tools for analyzing and interpreting neural activity. However, these approaches are unable to determine the neurophysiological meaning of the inferred latent dynamics. On the other hand, emerging evidence suggests that dynamic functional connectivities (DFC) may be responsible for neural activity patterns underlying cognition or behavior. We are interested in studying how DFC are associated with the low-dimensional structure of neural activities. Most existing LVMs are based on a point process and fail to model evolving relationships. In this work, we introduce a dynamic graph as the latent variable and develop a Variational Dynamic Graph Latent Variable Model (VDGLVM), a representation learning model based on the variational information bottleneck framework. VDGLVM utilizes a graph generative model and a graph neural network to capture dynamic communication between nodes that one has no access to from the observed data. The proposed computational model provides guaranteed behavior-decoding performance and improves LVMs by associating the inferred latent dynamics with probable DFC.<\/jats:p>","DOI":"10.3390\/e24020152","type":"journal-article","created":{"date-parts":[[2022,1,19]],"date-time":"2022-01-19T21:01:51Z","timestamp":1642626111000},"page":"152","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Representation Learning for Dynamic Functional Connectivities via Variational Dynamic Graph Latent Variable Models"],"prefix":"10.3390","volume":"24","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5925-0472","authenticated-orcid":false,"given":"Yicong","family":"Huang","sequence":"first","affiliation":[{"name":"College of Automation Science and Technology, South China University of Technology, Guangzhou 510641, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhuliang","family":"Yu","sequence":"additional","affiliation":[{"name":"College of Automation Science and Technology, South China University of Technology, Guangzhou 510641, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2022,1,19]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"232","DOI":"10.1038\/nature24636","article-title":"Fully integrated silicon probes for high-density recording of neural activity","volume":"551","author":"Jun","year":"2017","journal-title":"Nature"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"330","DOI":"10.1038\/s41583-019-0140-6","article-title":"Novel electrode technologies for neural recordings","volume":"20","author":"Hong","year":"2019","journal-title":"Nat. Rev. Neurosci."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"6539","DOI":"10.1126\/science.abf4588","article-title":"Neuropixels 2.0: A miniaturized high-density probe for stable, long-term brain recordings","volume":"372","author":"Steinmetz","year":"2021","journal-title":"Science"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"553","DOI":"10.1038\/s41592-019-0400-4","article-title":"High-density multi-fiber photometry for studying large-scale brain circuit dynamics","volume":"16","author":"Sych","year":"2019","journal-title":"Nat. Methods"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"415","DOI":"10.1016\/j.neuron.2012.01.010","article-title":"How does the brain solve visual object recognition?","volume":"73","author":"DiCarlo","year":"2012","journal-title":"Neuron"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"361","DOI":"10.1038\/s41586-019-1346-5","article-title":"High-dimensional geometry of population responses in visual cortex","volume":"571","author":"Stringer","year":"2019","journal-title":"Nature"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"1500","DOI":"10.1038\/nn.3776","article-title":"Dimensionality reduction for large-scale neural recordings","volume":"17","author":"Cunningham","year":"2014","journal-title":"Nat. Neurosci."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"805","DOI":"10.1038\/s41592-018-0109-9","article-title":"Inferring single-trial neural population dynamics using sequential auto-encoders","volume":"15","author":"Pandarinath","year":"2018","journal-title":"Nat. Methods"},{"key":"ref_9","first-page":"15937","article-title":"Enabling hyperparameter optimization in sequential autoencoders for spiking neural data","volume":"32","author":"Keshtkaran","year":"2019","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Ye, J., and Pandarinath, C. (2021). Representation learning for neural population activity with Neural Data Transformers. Neurons Behav. Data Anal. Theory.","DOI":"10.1101\/2021.01.16.426955"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Hurwitz, C., Kudryashova, N., Onken, A., and Hennig, M.H. (2021). Building population models for large-scale neural recordings: Opportunities and pitfalls. arXiv.","DOI":"10.1016\/j.conb.2021.07.003"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41467-018-06560-z","article-title":"Cortical population activity within a preserved neural manifold underlies multiple motor behaviors","volume":"9","author":"Gallego","year":"2018","journal-title":"Nat. Commun."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"964","DOI":"10.1016\/j.neuron.2018.09.030","article-title":"A neural population mechanism for rapid learning","volume":"100","author":"Perich","year":"2018","journal-title":"Neuron"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"672","DOI":"10.1038\/s41551-020-0542-9","article-title":"Stabilization of a brain\u2013computer interface via the alignment of low-dimensional spaces of neural activity","volume":"4","author":"Degenhart","year":"2020","journal-title":"Nat. Biomed. Eng."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"353","DOI":"10.1038\/nn.4502","article-title":"Network neuroscience","volume":"20","author":"Bassett","year":"2017","journal-title":"Nat. Neurosci."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"566","DOI":"10.1038\/s41583-018-0038-8","article-title":"On the nature and use of models in network neuroscience","volume":"19","author":"Bassett","year":"2018","journal-title":"Nat. Rev. Neurosci."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1385\/NI:2:2:205","article-title":"\u201cDynamic\u201d connectivity in neural systems","volume":"2","author":"Breakspear","year":"2004","journal-title":"Neuroinformatics"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"360","DOI":"10.1016\/j.neuroimage.2013.05.079","article-title":"Dynamic functional connectivity: Promise, issues, and interpretations","volume":"80","author":"Hutchison","year":"2013","journal-title":"Neuroimage"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"526","DOI":"10.1016\/j.neuroimage.2017.08.006","article-title":"Task-based dynamic functional connectivity: Recent findings and open questions","volume":"180","author":"Bandettini","year":"2018","journal-title":"Neuroimage"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"17","DOI":"10.1038\/nrn.2017.149","article-title":"Communication dynamics in complex brain networks","volume":"19","author":"Misic","year":"2018","journal-title":"Nat. Rev. Neurosci."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1109\/TNNLS.2020.2978386","article-title":"A comprehensive survey on graph neural networks","volume":"32","author":"Wu","year":"2020","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_22","unstructured":"Alemi, A.A., Fischer, I., Dillon, V.J., and Murphy, K. (2017, January 24\u201326). Deep Variational Information Bottleneck. Proceedings of the International Conference on Learning Representations, Toulon, France."},{"key":"ref_23","unstructured":"Byron, M.Y., Cunningham, J.P., Santhanam, G., Ryu, S.I., Shenoy, K.V., and Sahani, M. (2009, January 7\u201310). Gaussian-process factor analysis for low-dimensional single-trial analysis of neural population activity. Proceedings of the Advances in Neural Information Processing Systems 22 (NIPS 2009), Vancouver, BC, Canada."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1293","DOI":"10.1162\/NECO_a_00953","article-title":"Variational latent gaussian process for recovering single-trial dynamics from population spike trains","volume":"29","author":"Zhao","year":"2017","journal-title":"Neural Comput."},{"key":"ref_25","unstructured":"Wu, A., Roy, N.A., Keeley, S., and Pillow, J.W. (2017, January 4\u20139). Gaussian process based nonlinear latent structure discovery in multivariate spike train data. Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA."},{"key":"ref_26","unstructured":"She, Q., and Wu, A. (2020, January 3\u20136). Neural dynamics discovery via gaussian process recurrent neural networks. Proceedings of the 35th Uncertainty in Artificial Intelligence Conference, Virtual online."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Liu, D., and Lengyel, M. (2021, January 6\u201314). A universal probabilistic spike count model reveals ongoing modulation of neural variability. Proceedings of the Thirty-Fifth Conference on Neural Information Processing Systems, Online.","DOI":"10.1101\/2021.06.27.450063"},{"key":"ref_28","unstructured":"Macke, J.H., Buesing, L., Cunningham, J.P., Yu, B.M., Shenoy, K.V., and Sahani, M. (2011, January 12\u201315). Empirical models of spiking in neural populations. Proceedings of the Advances in Neural Information Processing Systems 24: 25th Conference on Neural Information Processing Systems (NIPS 2011), Granada, Spain."},{"key":"ref_29","first-page":"163","article-title":"Linear dynamical neural population models through nonlinear embeddings","volume":"29","author":"Gao","year":"2016","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Liu, R., Azabou, M., Dabagia, M., Lin, C.H., Gheshlaghi Azar, M., Hengen, K., Valko, M., and Dyer, E. (2021, January 6\u201314). Drop, Swap, and Generate: A Self-Supervised Approach for Generating Neural Activity. Proceedings of the Advances in Neural Information Processing Systems, Online.","DOI":"10.1101\/2021.07.21.453285"},{"key":"ref_31","first-page":"7234","article-title":"Learning identifiable and interpretable latent models of high-dimensional neural activity using pi-VAE","volume":"33","author":"Zhou","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"175","DOI":"10.3389\/fnsys.2015.00175","article-title":"A tutorial review of functional connectivity analysis methods and their interpretational pitfalls","volume":"9","author":"Bastos","year":"2016","journal-title":"Front. Syst. Neurosci."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1712","DOI":"10.1016\/j.neuroimage.2012.06.078","article-title":"Periodic changes in fMRI connectivity","volume":"63","author":"Handwerker","year":"2012","journal-title":"Neuroimage"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"3280","DOI":"10.1002\/hbm.22140","article-title":"Short-time windows of correlation between large-scale functional brain networks predict vigilance intraindividually and interindividually","volume":"34","author":"Thompson","year":"2013","journal-title":"Hum. Brain Mapp."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"10341","DOI":"10.1073\/pnas.1400181111","article-title":"Time-resolved resting-state brain networks","volume":"111","author":"Zalesky","year":"2014","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"242","DOI":"10.1016\/j.neuroimage.2015.11.055","article-title":"Can sliding-window correlations reveal dynamic functional connectivity in resting-state fMRI?","volume":"127","author":"Hindriks","year":"2016","journal-title":"Neuroimage"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"424","DOI":"10.2307\/1912791","article-title":"Investigating causal relations by econometric models and cross-spectral methods","volume":"37","author":"Granger","year":"1969","journal-title":"Econometrica"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"354","DOI":"10.1016\/j.neuroimage.2008.02.020","article-title":"Analyzing information flow in brain networks with nonparametric Granger causality","volume":"41","author":"Dhamala","year":"2008","journal-title":"Neuroimage"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"116796","DOI":"10.1016\/j.neuroimage.2020.116796","article-title":"Measuring directed functional connectivity using non-parametric directionality analysis: Validation and comparison with non-parametric Granger Causality","volume":"218","author":"West","year":"2020","journal-title":"NeuroImage"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"2379","DOI":"10.1007\/s10072-020-04759-x","article-title":"Dynamic functional connectivity in temporal lobe epilepsy: A graph theoretical and machine learning approach","volume":"42","author":"Fallahi","year":"2021","journal-title":"Neurol. Sci."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"97","DOI":"10.1016\/j.neucom.2021.05.003","article-title":"A deep autoencoder with sparse and graph Laplacian regularization for characterizing dynamic functional connectivity during brain development","volume":"456","author":"Qiao","year":"2021","journal-title":"Neurocomputing"},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Jiang, B., Huang, Y., Panahi, A., Yu, Y., Krim, H., and Smith, S.L. (2021). Dynamic Graph Learning: A Structure-Driven Approach. Mathematics, 9.","DOI":"10.3390\/math9020168"},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"119","DOI":"10.1007\/s10548-008-0071-4","article-title":"Characterizing dynamic functional connectivity across sleep stages from EEG","volume":"22","author":"Dimitriadis","year":"2009","journal-title":"Brain Topogr."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"101","DOI":"10.1007\/s10548-017-0546-2","article-title":"EEG signatures of dynamic functional network connectivity states","volume":"31","author":"Allen","year":"2018","journal-title":"Brain Topogr."},{"key":"ref_45","unstructured":"Gori, M., Monfardini, G., and Scarselli, F. (August, January 31). A new model for learning in graph domains. Proceedings of the 2005 IEEE International Joint Conference on Neural Networks, Montreal, QC, Canada."},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1109\/TNN.2008.2005605","article-title":"The graph neural network model","volume":"20","author":"Scarselli","year":"2008","journal-title":"IEEE Trans. Neural Netw."},{"key":"ref_47","unstructured":"Jin, W., Barzilay, R., and Jaakkola, T. (2018, January 10\u201315). Junction tree variational autoencoder for molecular graph generation. Proceedings of the 35th International Conference on Machine Learning, Stockholm, Sweden."},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"1038","DOI":"10.1016\/j.neuroimage.2016.09.046","article-title":"BrainNetCNN: Convolutional neural networks for brain networks; towards predicting neurodevelopment","volume":"146","author":"Kawahara","year":"2017","journal-title":"NeuroImage"},{"key":"ref_49","unstructured":"Gilmer, J., Schoenholz, S.S., Riley, P.F., Vinyals, O., and Dahl, G.E. (2017, January 6\u201311). Neural message passing for quantum chemistry. Proceedings of the 34th International Conference on Machine Learning, Sydney, Australia."},{"key":"ref_50","doi-asserted-by":"crossref","unstructured":"Tishby, N., and Zaslavsky, N. (May, January 26). Deep learning and the information bottleneck principle. Proceedings of the 2015 IEEE Information Theory Workshop (ITW), Jerusalem, Israel.","DOI":"10.1109\/ITW.2015.7133169"},{"key":"ref_51","unstructured":"Shwartz-Ziv, R., and Tishby, N. (2017). Opening the black box of deep neural networks via information. arXiv."},{"key":"ref_52","doi-asserted-by":"crossref","first-page":"1141","DOI":"10.1214\/aoms\/1177706098","article-title":"Random graphs","volume":"30","author":"Gilbert","year":"1959","journal-title":"Ann. Math. Stat."},{"key":"ref_53","unstructured":"Xu, K., Hu, W., Leskovec, J., and Jegelka, S. (May, January 30). How Powerful are Graph Neural Networks?. Proceedings of the International Conference on Learning Representations, Vancouver, BC, Canada."},{"key":"ref_54","doi-asserted-by":"crossref","unstructured":"Chepuri, S.P., Liu, S., Leus, G., and Hero, A.O. (2017, January 5\u20139). Learning sparse graphs under smoothness prior. Proceedings of the 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, LA, USA.","DOI":"10.1109\/ICASSP.2017.7953410"},{"key":"ref_55","unstructured":"Pei, F.C., Ye, J., Zoltowski, D.M., Wu, A., Chowdhury, R.H., Sohn, H., O\u2019Doherty, J.E., Shenoy, K.V., Kaufman, M., and Churchland, M.M. (2021, January 6\u201314). Neural Latents Benchmark \u201821: Evaluating latent variable models of neural population activity. Proceedings of the Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2), Virtual."},{"key":"ref_56","doi-asserted-by":"crossref","first-page":"e48198","DOI":"10.7554\/eLife.48198","article-title":"Area 2 of primary somatosensory cortex encodes kinematics of the whole arm","volume":"9","author":"Chowdhury","year":"2020","journal-title":"eLife"},{"key":"ref_57","doi-asserted-by":"crossref","first-page":"387","DOI":"10.1016\/j.neuron.2010.09.015","article-title":"Cortical preparatory activity: Representation of movement or first cog in a dynamical machine?","volume":"68","author":"Churchland","year":"2010","journal-title":"Neuron"}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/24\/2\/152\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T22:04:15Z","timestamp":1760133855000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/24\/2\/152"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,1,19]]},"references-count":57,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2022,2]]}},"alternative-id":["e24020152"],"URL":"https:\/\/doi.org\/10.3390\/e24020152","relation":{},"ISSN":["1099-4300"],"issn-type":[{"type":"electronic","value":"1099-4300"}],"subject":[],"published":{"date-parts":[[2022,1,19]]}}}