pith. sign in

arxiv: 2606.22765 · v1 · pith:PGSPERWJnew · submitted 2026-06-22 · 💻 cs.NE · cs.AI· cs.LG· nlin.CD· physics.comp-ph

Evolutionary Optimization Reveals Structural Constraints on Reservoir Architecture for Spatiotemporal Chaos

Nima Dehghani This is my paper

Pith reviewed 2026-06-26 06:16 UTC · model grok-4.3

classification 💻 cs.NE cs.AIcs.LGnlin.CDphysics.comp-ph
keywords reservoir computingevolutionary optimizationKuramoto-Sivashinsky equationspatiotemporal chaosstructural constraintsecho state networksprediction horizon
0
0 comments X

The pith

Evolutionary selection on reservoir hyperparameters exposes conserved structural constraints that improve prediction of spatiotemporal chaos.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper places the recurrent substrate of reservoir computers under evolutionary pressure for forecasting the Kuramoto-Sivashinsky equation and tracks how the five construction hyperparameters change across generations. Successful networks converge on a shared spectral envelope resembling a stochastic block model, refine low-eigenvalue modes, and stabilize modularity within a narrow intermediate band while pruning connection cost. Accuracy and efficiency are achieved jointly on a horizontal cost-modularity floor rather than through a trade-off. A sympathetic reader would care because the results suggest that task demands can derive interpretable architectural rules for recurrent networks instead of relying on random initialization.

Core claim

When five reservoir hyperparameters (size, connectivity degree, spectral radius, input scaling, readout regularization) are evolved for Kuramoto-Sivashinsky prediction, the resulting architectures remain inside a conserved stochastic-block-model-like spectral envelope, refine low-eigenvalue modes, lock modularity to an intermediate band, and occupy a horizontal floor in the cost-modularity plane where prediction accuracy and efficiency improve together.

What carries the argument

Evolutionary optimization over the five reservoir construction hyperparameters, which selects networks for prediction performance and thereby stabilizes a task-suitable dynamical class.

If this is right

  • Evolved reservoirs achieve lower prediction error and longer low-error forecast horizons than randomly initialized ones.
  • The design space organizes along a diminishing-return size-efficiency frontier.
  • Elite networks jointly optimize accuracy and connection cost rather than trading one for the other.
  • Structural refinement occurs mainly by pruning within a conserved spectral and modular band.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same evolutionary procedure could be run on other chaotic partial differential equations to test whether the spectral-envelope and modularity constraints are system-specific or task-general.
  • If the intermediate modularity band proves robust, it could serve as a prior when designing recurrent networks for other forecasting tasks.
  • Varying the fitness function (for example, adding an explicit energy cost) would likely shift the location of the cost-modularity floor and reveal different structural constraints.

Load-bearing premise

The assumption that five evolved hyperparameters on a single Kuramoto-Sivashinsky testbed suffice to reveal general structural constraints that apply to other chaotic systems or fitness functions.

What would settle it

Reservoirs evolved on a different spatiotemporal chaotic system, such as the two-dimensional Navier-Stokes equations, that fail to converge on the same spectral envelope or intermediate modularity band would falsify the claimed generality of the constraints.

Figures

Figures reproduced from arXiv: 2606.22765 by Nima Dehghani.

Figure 1
Figure 1. Figure 1: FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7 [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8 [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9 [PITH_FULL_IMAGE:figures/full_fig_p010_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10 [PITH_FULL_IMAGE:figures/full_fig_p011_10.png] view at source ↗
read the original abstract

Biological systems maintain function in fluctuating environments by transforming past stimulation into internal dynamical states that support future-oriented responses. Reservoir computing provides a computational analogue, but standard formulations often treat the recurrent substrate as a fixed random network and train only the readout. Here we ask how the substrate itself changes when reservoir architecture is placed under evolutionary selection for prediction. Using the Kuramoto--Sivashinsky equation as a testbed for spatiotemporal chaos, we evolved reservoirs over five construction hyperparameters: size, connectivity degree, spectral radius, input scaling, and readout regularization. Evolution reduced prediction error at the population level, extended the low-error forecast horizon, and organized the design space along a diminishing-return size--efficiency frontier. Structural analyses showed that evolved reservoirs remained within a conserved stochastic-block-model-like spectral envelope while refining low-eigenvalue modes, locking modularity to an intermediate band, and pruning connection cost within that band. Pareto analysis showed that elite reservoirs occupied a horizontal floor in the cost--modularity plane, indicating that accuracy and efficiency were achieved jointly rather than through a simple trade-off. These findings show that evolutionary optimization does not merely improve prediction, but exposes interpretable structural constraints on the recurrent substrate: it stabilizes a task-suitable dynamical class and refines the architectural degrees of freedom most relevant for prediction. Evolutionary reservoir computing therefore provides a bio-inspired framework for studying how predictive demands shape adaptive dynamical networks.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper claims that placing reservoir architecture under evolutionary selection for prediction on the Kuramoto-Sivashinsky equation (using five hyperparameters: size, connectivity degree, spectral radius, input scaling, readout regularization) yields population-level error reduction, extended forecast horizons, a conserved stochastic-block-model-like spectral envelope with refined low-eigenvalue modes, an intermediate modularity band, connection-cost pruning, and a horizontal Pareto floor in the cost-modularity plane. It concludes that evolution stabilizes a task-suitable dynamical class and exposes interpretable structural constraints on the recurrent substrate beyond mere performance gains.

Significance. If the structural findings hold and generalize, the work would provide a bio-inspired method for identifying how predictive demands constrain reservoir dynamics and architecture, moving beyond random substrates to reveal refined degrees of freedom. The direct evolutionary simulation on an external dynamical system (rather than parameter fitting) is a strength, as is the joint accuracy-efficiency outcome on the Pareto front. However, the single-testbed design limits the reach of the 'general structural constraints' interpretation.

major comments (2)
  1. [Abstract (final paragraph)] Abstract (final paragraph) and implied Results: The central claim that evolution 'exposes interpretable structural constraints on the recurrent substrate' (conserved spectral envelope, intermediate modularity band, Pareto floor) is load-bearing for the paper's contribution, yet all analyses use a single spatiotemporal chaotic system (Kuramoto-Sivashinsky) and the same five hyperparameters with no cross-system validation or alternative fitness functions reported. This makes it impossible to determine whether the observed features are general properties of predictive reservoirs or KS-specific artifacts.
  2. [Abstract] Abstract: The reported population-level error reduction, extended low-error forecast horizon, and structural analyses provide no quantitative details on number of independent evolutionary runs, statistical tests, baseline comparisons (e.g., random vs. evolved), or exclusion criteria, undermining assessment of whether the claimed refinements are robust or merely descriptive of the single run set.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed comments. We agree that the abstract overstates the generality of the structural findings and lacks explicit quantitative reporting. Both issues will be addressed through targeted revisions to the abstract and addition of a limitations paragraph.

read point-by-point responses

  1. Referee: Abstract (final paragraph) and implied Results: The central claim that evolution 'exposes interpretable structural constraints on the recurrent substrate' (conserved spectral envelope, intermediate modularity band, Pareto floor) is load-bearing for the paper's contribution, yet all analyses use a single spatiotemporal chaotic system (Kuramoto-Sivashinsky) and the same five hyperparameters with no cross-system validation or alternative fitness functions reported. This makes it impossible to determine whether the observed features are general properties of predictive reservoirs or KS-specific artifacts.

    Authors: We accept that the single-testbed design (Kuramoto-Sivashinsky only) precludes claims of general structural constraints across predictive reservoirs. The final paragraph of the abstract will be revised to state that the conserved spectral envelope, intermediate modularity, and Pareto floor are observed under evolutionary selection for this specific spatiotemporal chaotic system. A new limitations subsection will be added to the discussion explicitly noting the absence of cross-system validation and alternative fitness functions, framing the results as a detailed case study rather than a general principle. revision: yes

  2. Referee: Abstract: The reported population-level error reduction, extended low-error forecast horizon, and structural analyses provide no quantitative details on number of independent evolutionary runs, statistical tests, baseline comparisons (e.g., random vs. evolved), or exclusion criteria, undermining assessment of whether the claimed refinements are robust or merely descriptive of the single run set.

    Authors: The full manuscript reports these quantities in the Methods (evolutionary algorithm parameters and run count) and Results (statistical comparisons to random baselines and exclusion of divergent simulations), but the abstract does not summarize them. We will add a concise quantitative statement to the abstract specifying the number of independent evolutionary runs, the statistical tests employed, the random baseline comparisons, and the exclusion criteria. revision: yes

Circularity Check

0 steps flagged

No significant circularity; results from direct evolutionary simulation on external system

full rationale

The paper's central results derive from running evolutionary optimization on five reservoir hyperparameters against the Kuramoto-Sivashinsky equation as an external testbed, followed by post-hoc structural analysis of the evolved networks. No derivation step reduces by construction to fitted inputs, self-definitions, or self-citation chains; the observed spectral envelopes, modularity bands, and Pareto floors are outputs of the simulation rather than reparameterizations of the optimization objective. The single-testbed limitation affects generalizability but does not create circularity in the reported chain.

Axiom & Free-Parameter Ledger

5 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard reservoir-computing assumptions plus the choice of the Kuramoto-Sivashinsky equation as representative of spatiotemporal chaos; no new entities are postulated.

free parameters (5)
  • reservoir size
    One of five hyperparameters placed under evolutionary selection
  • connectivity degree
    One of five hyperparameters placed under evolutionary selection
  • spectral radius
    One of five hyperparameters placed under evolutionary selection
  • input scaling
    One of five hyperparameters placed under evolutionary selection
  • readout regularization
    One of five hyperparameters placed under evolutionary selection
axioms (2)
  • domain assumption The Kuramoto-Sivashinsky equation is a suitable testbed for spatiotemporal chaos
    Used as the sole prediction task throughout the abstract
  • domain assumption Standard reservoir computing assumptions (fixed recurrent weights, trainable readout, echo-state-like behavior) hold for the evolved networks
    Implicit in the use of reservoir computing as the substrate

pith-pipeline@v0.9.1-grok · 5787 in / 1621 out tokens · 29919 ms · 2026-06-26T06:16:12.462673+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

106 extracted references · 70 canonical work pages

  1. [1]

    Predictive Behavior Within Microbial Genetic Networks

    Tagkopoulos, Ilias and Liu, Yir-Chung and Tavazoie, Saeed. Predictive Behavior Within Microbial Genetic Networks. Science. 2008. doi:10.1126/science.1154456

    work page doi:10.1126/science.1154456 2008

  2. [2]

    Frontiers in Physiology , VOLUME=

    Deans, Carrie , TITLE=. Frontiers in Physiology , VOLUME=. 2021 , URL=. doi:10.3389/fphys.2021.672457 , ISSN=

    work page doi:10.3389/fphys.2021.672457 2021

  3. [3]

    and Tavazoie, Saeed

    Freddolino, Peter L. and Tavazoie, Saeed. Beyond Homeostasis: A Predictive-Dynamic Framework for Understanding Cellular Behavior. Annual Review of Cell and Developmental Biology. 2012. doi:10.1146/annurev-cellbio-092910-154129

    work page doi:10.1146/annurev-cellbio-092910-154129 2012

  4. [4]

    and Garcia-Ojalvo, Jordi

    Gabalda-Sagarra, Marçal and Carey, Lucas B. and Garcia-Ojalvo, Jordi. Recurrence-based information processing in gene regulatory networks. Chaos: An Interdisciplinary Journal of Nonlinear Science. 2018. doi:10.1063/1.5039861

    work page doi:10.1063/1.5039861 2018

  5. [5]

    Echo State Property of Deep Reservoir Computing Networks

    Gallicchio, Claudio and Micheli, Alessio. Echo State Property of Deep Reservoir Computing Networks. Cognitive Computation. 2017. doi:10.1007/s12559-017-9461-9

    work page doi:10.1007/s12559-017-9461-9 2017

  6. [6]

    Model-Free Prediction of Large Spatiotemporally Chaotic Systems from Data: A Reservoir Computing Approach , author =. Phys. Rev. Lett. , volume =. 2018 , month =. doi:10.1103/PhysRevLett.120.024102 , url =

    work page doi:10.1103/physrevlett.120.024102 2018

  7. [7]

    and Girvan, Michelle and Ott, Edward , title =

    Pathak, Jaideep and Lu, Zhixin and Hunt, Brian R. and Girvan, Michelle and Ott, Edward , title =. Chaos: An Interdisciplinary Journal of Nonlinear Science , volume =. 2017 , month =. doi:10.1063/1.5010300 , url =

    work page doi:10.1063/1.5010300 2017

  8. [8]

    and Girvan, Michelle and Ott, Edward , title =

    Pathak, Jaideep and Wikner, Alexander and Fussell, Rebeckah and Chandra, Sarthak and Hunt, Brian R. and Girvan, Michelle and Ott, Edward , title =. Chaos: An Interdisciplinary Journal of Nonlinear Science , volume =. 2018 , month =. doi:10.1063/1.5028373 , url =

    work page doi:10.1063/1.5028373 2018

  9. [9]

    Tomlin, R. J. and Kalogirou, A. and Papageorgiou, D. T. , title =. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences , volume =. 2018 , month =. doi:10.1098/rspa.2017.0687 , url =

    work page doi:10.1098/rspa.2017.0687 2018

  10. [10]

    A particle model for the Kuramoto-Sivashinsky equation

    Rost, Martin and Krug, Joachim. A particle model for the Kuramoto-Sivashinsky equation. Physica D: Nonlinear Phenomena. 1995. doi:10.1016/0167-2789(95)00184-6

    work page doi:10.1016/0167-2789(95)00184-6 1995

  11. [11]

    , title =

    Seoane, Luís F. , title =. Philosophical Transactions of the Royal Society B: Biological Sciences , volume =. 2019 , month =. doi:10.1098/rstb.2018.0377 , url =

    work page doi:10.1098/rstb.2018.0377 2019

  12. [12]

    Good and bad predictions: Assessing and improving the replication of chaotic attractors by means of reservoir computing

    Haluszczynski, Alexander and Räth, Christoph. Good and bad predictions: Assessing and improving the replication of chaotic attractors by means of reservoir computing. Chaos: An Interdisciplinary Journal of Nonlinear Science. 2019. doi:10.1063/1.5118725

    work page doi:10.1063/1.5118725 2019

  13. [13]

    small-world

    Kitayama, Ken-ichi. Guiding principle of reservoir computing based on “small-world” network. Scientific Reports. 2022. doi:10.1038/s41598-022-21235-y

    work page doi:10.1038/s41598-022-21235-y 2022

  14. [14]

    Complex topological features of reservoirs shape learning performances in bio-inspired recurrent neural networks

    d'Andrea, Valeria and Puppin, Michele and Domenico, Manlio De. Complex topological features of reservoirs shape learning performances in bio-inspired recurrent neural networks. Preprint. 2211. doi:10.48550/arxiv.2211.00161

    work page doi:10.48550/arxiv.2211.00161

  15. [15]

    and Menzinger, Michael

    McGraw, Patrick N. and Menzinger, Michael. Laplacian spectra as a diagnostic tool for network structure and dynamics. Physical Review E. 2008. doi:10.1103/physreve.77.031102

    work page doi:10.1103/physreve.77.031102 2008

  16. [16]

    echo state

    Jaeger, Herbert , year=. The “echo state” approach to analysing and training recurrent neural networks , url=. doi:10.24406/publica-fhg-291111 , publisher=

    work page doi:10.24406/publica-fhg-291111

  17. [17]

    Neural Computation , volume =

    Maass, Wolfgang and Natschläger, Thomas and Markram, Henry , title =. Neural Computation , volume =. 2002 , month =. doi:10.1162/089976602760407955 , url =

    work page doi:10.1162/089976602760407955 2002

  18. [18]

    Reservoir Computing Approaches to Recurrent Neural Network Training , journal =

    Luko. Reservoir Computing Approaches to Recurrent Neural Network Training , journal =. 2009 , volume =

    2009

  19. [19]

    Current Opinion in Neurobiology , year =

    Sussillo, David , title =. Current Opinion in Neurobiology , year =

  20. [20]

    A Survey on Reservoir Computing and its Interdisciplinary Applications Beyond Traditional Machine Learning , year=

    Zhang, Heng and Vargas, Danilo Vasconcellos , journal=. A Survey on Reservoir Computing and its Interdisciplinary Applications Beyond Traditional Machine Learning , year=

  21. [21]

    Logic Journal of the IGPL , volume =

    Roeschies, Benjamin and Igel, Christian , title =. Logic Journal of the IGPL , volume =. 2010 , month =. doi:10.1093/jigpal/jzp043 , url =

    work page doi:10.1093/jigpal/jzp043 2010

  22. [22]

    and Xu, Dongming and Príncipe, José C

    Ozturk, Mustafa C. and Xu, Dongming and Príncipe, José C. Analysis and Design of Echo State Networks. Neural Computation. 2007. doi:10.1162/neco.2007.19.1.111

    work page doi:10.1162/neco.2007.19.1.111 2007

  23. [23]

    Improving the prediction of spatio-temporal chaos by combining parallel reservoir computing with dimensionality reduction

    Fleddermann, Luk and Parlitz, Ulrich and Wellecke, Gerrit. Improving the prediction of spatio-temporal chaos by combining parallel reservoir computing with dimensionality reduction. Machine Learning: Science and Technology. 2025. doi:10.1088/2632-2153/ae1806

    work page doi:10.1088/2632-2153/ae1806 2025

  24. [24]

    Effects of singular value spectrum on the performance of echo state network

    Li, Fanjun and Wang, Xiaohong and Li, Ying. Effects of singular value spectrum on the performance of echo state network. Neurocomputing. 2019. doi:10.1016/j.neucom.2019.05.068

    work page doi:10.1016/j.neucom.2019.05.068 2019

  25. [25]

    Neuronal timescales are functionally dynamic and shaped by cortical microarchitecture

    Gao, Richard and van den Brink, Ruud L and Pfeffer, Thomas and Voytek, Bradley. Neuronal timescales are functionally dynamic and shaped by cortical microarchitecture. eLife. 2020. doi:10.7554/elife.61277

    work page doi:10.7554/elife.61277 2020

  26. [26]

    Cell , year=

    Chen, Guang and Kang, Byungwoo and Lindsey, Jack and Druckmann, Shaul and Li, Nuo , title=. Cell , year=. doi:10.1016/j.cell.2021.05.026 , url=

    work page doi:10.1016/j.cell.2021.05.026 2021

  27. [27]

    Modularity in the Organization of Mouse Primary Visual Cortex , journal=

    Ji, Weiqing and G. Modularity in the Organization of Mouse Primary Visual Cortex , journal=. 2015 , month=. doi:10.1016/j.neuron.2015.07.004 , url=

    work page doi:10.1016/j.neuron.2015.07.004 2015

  28. [28]

    Powell and Bettina Hein and Deyue Kong and Jonas Elpelt and Haleigh N

    Nathaniel J. Powell and Bettina Hein and Deyue Kong and Jonas Elpelt and Haleigh N. Mulholland and Matthias Kaschube and Gordon B. Smith , title =. Proceedings of the National Academy of Sciences , volume =. 2024 , doi =

    2024

  29. [29]

    Modular Brain Networks

    Sporns, Olaf and Betzel, Richard F. Modular Brain Networks. Annual review of psychology. 2016. doi:10.1146/annurev-psych-122414-033634

    work page doi:10.1146/annurev-psych-122414-033634 2016

  30. [30]

    , TITLE=

    Meunier, David and Lambiotte, Renaud and Bullmore, Edward T. , TITLE=. Frontiers in Neuroscience , VOLUME=. 2010 , URL=. doi:10.3389/fnins.2010.00200 , ISSN=

    work page doi:10.3389/fnins.2010.00200 2010

  31. [31]

    The Wiring Economy Principle: Connectivity Determines Anatomy in the Human Brain

    Raj, Ashish and Chen, Yu-hsien. The Wiring Economy Principle: Connectivity Determines Anatomy in the Human Brain. PLoS ONE. 2011. doi:10.1371/journal.pone.0014832

    work page doi:10.1371/journal.pone.0014832 2011

  32. [32]

    and Clandinin, Thomas R

    Wang, Irving E. and Clandinin, Thomas R. The Influence of Wiring Economy on Nervous System Evolution. Current Biology. 2016. doi:10.1016/j.cub.2016.08.053

    work page doi:10.1016/j.cub.2016.08.053 2016

  33. [33]

    Budd, Julian M. L. and Kisvárday, Zoltán F. Communication and wiring in the cortical connectome. Frontiers in Neuroanatomy. 2012. doi:10.3389/fnana.2012.00042

    work page doi:10.3389/fnana.2012.00042 2012

  34. [34]

    and Daunizeau, Jean and Friston, Karl J

    Kiebel, Stefan J. and Daunizeau, Jean and Friston, Karl J. A Hierarchy of Time-Scales and the Brain. PLoS Computational Biology. 2008. doi:10.1371/journal.pcbi.1000209

    work page doi:10.1371/journal.pcbi.1000209 2008

  35. [35]

    A Large-Scale Circuit Mechanism for Hierarchical Dynamical Processing in the Primate Cortex

    Chaudhuri, Rishidev and Knoblauch, Kenneth and Gariel, Marie-Alice and Kennedy, Henry and Wang, Xiao-Jing. A Large-Scale Circuit Mechanism for Hierarchical Dynamical Processing in the Primate Cortex. Neuron. 2015. doi:10.1016/j.neuron.2015.09.008

    work page doi:10.1016/j.neuron.2015.09.008 2015

  36. [36]

    and Snyder, Abraham Z

    Raut, Ryan V. and Snyder, Abraham Z. and Raichle, Marcus E. Hierarchical dynamics as a macroscopic organizing principle of the human brain. Proceedings of the National Academy of Sciences. 2020. doi:10.1073/pnas.2003383117

    work page doi:10.1073/pnas.2003383117 2020

  37. [37]

    and Sheftel, H

    Shoval, O. and Sheftel, H. and Shinar, G. and Hart, Y. and Ramote, O. and Mayo, A. and Dekel, E. and Kavanagh, K. and Alon, U. Evolutionary Trade-Offs, Pareto Optimality, and the Geometry of Phenotype Space. Science. 2012. doi:10.1126/science.1217405

    work page doi:10.1126/science.1217405 2012

  38. [38]

    2021 , note =

    Neural optimization: Understanding trade-offs with Pareto theory , journal =. 2021 , note =. doi:10.1016/j.conb.2021.08.008 , url =

    work page doi:10.1016/j.conb.2021.08.008 2021

  39. [39]

    Pareto Optimality across different scales

    Szekely, Pablo. Pareto Optimality across different scales. Text. 2020. doi:10.34933/wis.000185

    work page doi:10.34933/wis.000185 2020

  40. [40]

    Variability in Neural Circuit Formation

    Mitchell, Kevin J. Variability in Neural Circuit Formation. Cold Spring Harbor perspectives in biology. 2024. doi:10.1101/cshperspect.a041504

    work page doi:10.1101/cshperspect.a041504 2024

  41. [41]

    Architectures of neuronal circuits

    Luo, Liqun. Architectures of neuronal circuits. Science. 2021. doi:10.1126/science.abg7285

    work page doi:10.1126/science.abg7285 2021

  42. [42]

    and Barabási, Albert-László

    Barabási, Dániel L. and Barabási, Albert-László. A Genetic Model of the Connectome. Neuron. 2020. doi:10.1016/j.neuron.2019.10.031

    work page doi:10.1016/j.neuron.2019.10.031 2020

  43. [43]

    Proceedings of the National Academy of Sciences , volume =

    Oren Richter and Elad Schneidman , title =. Proceedings of the National Academy of Sciences , volume =. 2025 , doi =

    2025

  44. [44]

    PLOS Computational Biology , publisher =

    Constructive connectomics: How neuronal axons get from here to there using gene-expression maps derived from their family trees , year =. PLOS Computational Biology , publisher =. doi:10.1371/journal.pcbi.1010382 , author =

    work page doi:10.1371/journal.pcbi.1010382

  45. [45]

    Brain-inspired modular echo state network for EEG-based emotion recognition

    Yang, Liuyi and Wang, Zhaoze and Wang, Guoyu and Liang, Lixin and Liu, Meng and Wang, Junsong. Brain-inspired modular echo state network for EEG-based emotion recognition. Frontiers in Neuroscience. 2024. doi:10.3389/fnins.2024.1305284

    work page doi:10.3389/fnins.2024.1305284 2024

  46. [46]

    and Goulas, Alexandros

    Damicelli, Fabrizio and Hilgetag, Claus C. and Goulas, Alexandros. Brain connectivity meets reservoir computing. PLOS Computational Biology. 2022. doi:10.1371/journal.pcbi.1010639

    work page doi:10.1371/journal.pcbi.1010639 2022

  47. [47]

    Accessing the topological properties of human brain functional sub-circuits in Echo State Networks

    Nguyen, Bach and Chen, Tianlong and Yang, Shu and Hou, Bojian and Shen, Li and Duong-Tran, Duy. Accessing the topological properties of human brain functional sub-circuits in Echo State Networks. 2024. doi:10.48550/arxiv.2412.14999

    work page doi:10.48550/arxiv.2412.14999 2024

  48. [48]

    and Gao, Richard

    Zeraati, Roxana and Levina, Anna and Macke, Jakob H. and Gao, Richard. Neural timescales from a computational perspective. 2409. doi:10.48550/arxiv.2409.02684

    work page doi:10.48550/arxiv.2409.02684

  49. [49]

    Frontiers in Computational Neuroscience , year =

    Dehghani, Nima , title =. Frontiers in Computational Neuroscience , year =

  50. [50]

    Journal of Physics: Complexity , year =

    Dehghani, Nima and Caterina, Gianluca , title =. Journal of Physics: Complexity , year =

  51. [51]

    PLOS Complex Systems , year =

    Dehghani, Nima , title =. PLOS Complex Systems , year =

  52. [52]

    Proceedings of the Royal Society B: Biological Sciences , year =

    Clune, Jeff and Mouret, Jean-Baptiste and Lipson, Hod , title =. Proceedings of the Royal Society B: Biological Sciences , year =

  53. [53]

    PLOS Computational Biology , year =

    Ellefsen, Kai Olav and Mouret, Jean-Baptiste and Clune, Jeff , title =. PLOS Computational Biology , year =

  54. [54]

    , title =

    Carroll, Thomas L. , title =. Chaos: An Interdisciplinary Journal of Nonlinear Science , year =

  55. [55]

    and Pecora, Louis M

    Carroll, Thomas L. and Pecora, Louis M. , title =. Chaos: An Interdisciplinary Journal of Nonlinear Science , year =

  56. [56]

    and Parlitz, Ulrich , title =

    Zimmermann, Roland S. and Parlitz, Ulrich , title =. Chaos: An Interdisciplinary Journal of Nonlinear Science , year =

  57. [57]

    , title =

    Niraula, Dipesh and El Naqa, Issam and Tuszynski, Jack Adam and Gatenby, Robert A. , title =. iScience , year =

  58. [58]

    and de Jong, Thijs J

    Baltussen, Mathieu G. and de Jong, Thijs J. and Duez, Quentin C. N. and Robinson, William E. and Huck, Wilhelm T. S. , title =. Nature , year =

  59. [59]

    2026 , eprint=

    What Makes a Bacterial Model a Good Reservoir Computer? Predicting Performance from Separability and Similarity , author=. 2026 , eprint=

    2026

  60. [60]

    and Pratap, A

    Deb, K. and Pratap, A. and Agarwal, S. and Meyarivan, T. , journal=. A fast and elitist multiobjective genetic algorithm:. 2002 , volume=

    2002

  61. [61]

    2022 , publisher =

    Jesús-Adolfo Mejía-de-Dios and Efrén Mezura-Montes , title =. 2022 , publisher =. doi:10.21105/joss.04723 , url =

    work page doi:10.21105/joss.04723 2022

  62. [62]

    Progress of Theoretical Physics Supplement , volume =

    Kuramoto, Yoshiki , title =. Progress of Theoretical Physics Supplement , volume =. 1978 , month =. doi:10.1143/PTPS.64.346 , url =

    work page doi:10.1143/ptps.64.346 1978

  63. [63]

    Derivation of basic equations , journal =

    Nonlinear analysis of hydrodynamic instability in laminar flames—I. Derivation of basic equations , journal =. 1977 , issn =. doi:10.1016/0094-5765(77)90096-0 , url =

    work page doi:10.1016/0094-5765(77)90096-0 1977

  64. [64]

    and Lodi, Andrea , title=

    Tanneau, Mathieu and Anjos, Miguel F. and Lodi, Andrea , title=. Mathematical Programming Computation , year=. doi:10.1007/s12532-020-00200-8 , url=

    work page doi:10.1007/s12532-020-00200-8

  65. [65]

    JuliaOptimalTransport/OptimalTransport.jl , year =

    Zhang, Stephen and Widmann, David and Barreira, Davi Sales and Besan. JuliaOptimalTransport/OptimalTransport.jl , year =

  66. [66]

    , title =

    Rubner, Yossi and Tomasi, Carlo and Guibas, Leonidas J. , title =. International Journal of Computer Vision , year =

  67. [67]

    Computational Optimal Transport , journal =

    Peyr. Computational Optimal Transport , journal =. 2019 , volume =

    2019

  68. [68]

    Newman, M. E. J. and Girvan, M. , title =. Physical Review E , year =

  69. [69]

    Statistical Mechanics of Community Detection , journal =

    Reichardt, J. Statistical Mechanics of Community Detection , journal =. 2006 , volume =

    2006

  70. [70]

    Leicht, E. A. and Newman, M. E. J. , title =. Physical Review Letters , year =

  71. [71]

    Near Linear Time Algorithm to Detect Community Structures in Large-Scale Networks , journal =

    Raghavan, Usha Nandini and Albert, R. Near Linear Time Algorithm to Detect Community Structures in Large-Scale Networks , journal =. 2007 , volume =

    2007

  72. [72]

    JuliaGraphs/Graphs.jl: an optimized graphs package for the Julia programming language , year =

    Fairbanks, James and Besan. JuliaGraphs/Graphs.jl: an optimized graphs package for the Julia programming language , year =

  73. [73]

    1997 , publisher=

    Spectral Graph Theory , author=. 1997 , publisher=

    1997

  74. [74]

    Annals of Combinatorics , year=

    Chung, Fan , title=. Annals of Combinatorics , year=. doi:10.1007/s00026-005-0237-z , url=

    work page doi:10.1007/s00026-005-0237-z

  75. [75]

    and de Reus, Marcel A

    de Lange, Siemon C. and de Reus, Marcel A. and van den Heuvel, Martijn P. The Laplacian spectrum of neural networks. Frontiers Comput. Neurosci. 2014. doi:10.3389/fncom.2013.00189

    work page doi:10.3389/fncom.2013.00189 2014

  76. [76]

    and Sorensen, Danny C

    Lehoucq, Richard B. and Sorensen, Danny C. and Yang, Chao , title =. 1998 , doi =

    1998

  77. [77]

    2002 , issn =

    Exponential Time Differencing for Stiff Systems , journal =. 2002 , issn =. doi:10.1006/jcph.2002.6995 , url =

    work page doi:10.1006/jcph.2002.6995 2002

  78. [78]

    , title =

    Kassam, Aly-Khan and Trefethen, Lloyd N. , title =. SIAM Journal on Scientific Computing , volume =. 2005 , doi =

    2005

  79. [79]

    Neuron , volume = 34, number = 3, pages =

    Wiring Optimization in Cortical Circuits , author =. Neuron , volume = 34, number = 3, pages =. doi:10.1016/S0896-6273(02)00679-7 , issn =

    work page doi:10.1016/s0896-6273(02)00679-7

  80. [80]

    Chen and David H

    Beth L. Chen and David H. Hall and Dmitri B. Chklovskii , title =. Proceedings of the National Academy of Sciences , volume =. 2006 , doi =

    2006

Showing first 80 references.