LOOPER: Inferring computational algorithms enacted by neuronal population dynamics

Recording simultaneous activity of hundreds of neurons is now possible. Existing methods can model such population activity, but do not directly reveal the computations used by the brain. We present a fully unsupervised method that models neuronal activity and reveals the computational strategy. The method constructs a topological model of neuronal dynamics consisting of interconnected loops. Transitions between loops mark computationally-salient decisions. We accurately model activation of 100s of neurons in the primate cortex during a working memory task. Dynamics of a recurrent neural network (RNN) trained on the same task are topologically identical suggesting that a similar computational strategy is used. The RNN trained on a modified dataset, however, reveals a different topology. This topology predicts specific novel stimuli that consistently elicit incorrect responses with near perfect accuracy. Thus, our methodology yields a quantitative model of neuronal activity and reveals the computational strategy used to solve the task.

[1]  Amit Singer,et al.  Detecting intrinsic slow variables in stochastic dynamical systems by anisotropic diffusion maps , 2009, Proceedings of the National Academy of Sciences.

[2]  Surya Ganguli,et al.  A theory of multineuronal dimensionality, dynamics and measurement , 2017, bioRxiv.

[3]  Rishidev Chaudhuri,et al.  The intrinsic attractor manifold and population dynamics of a canonical cognitive circuit across waking and sleep , 2019, Nature Neuroscience.

[4]  Konrad Kording,et al.  Annals of the New York Academy of Sciences Bayesian Models: the Structure of the World, Uncertainty, Behavior, and the Brain , 2022 .

[5]  Theodore H. Lindsay,et al.  Global Brain Dynamics Embed the Motor Command Sequence of Caenorhabditis elegans , 2015, Cell.

[6]  G. Laurent,et al.  Odor encoding as an active, dynamical process: experiments, computation, and theory. , 2001, Annual review of neuroscience.

[7]  S T Roweis,et al.  Nonlinear dimensionality reduction by locally linear embedding. , 2000, Science.

[8]  Balázs Rózsa,et al.  Fast two-photon in vivo imaging with three-dimensional random-access scanning in large tissue volumes , 2012, Nature Methods.

[9]  Peter Ford Dominey,et al.  Reservoir Computing Properties of Neural Dynamics in Prefrontal Cortex , 2016, PLoS Comput. Biol..

[10]  Byron M. Yu,et al.  Single-Trial Neural Correlates of Arm Movement Preparation , 2011, Neuron.

[11]  Jascha Sohl-Dickstein,et al.  SVCCA: Singular Vector Canonical Correlation Analysis for Deep Learning Dynamics and Interpretability , 2017, NIPS.

[12]  David W. Tank,et al.  Probing variability in a cognitive map using manifold inference from neural dynamics , 2018, bioRxiv.

[13]  N. Parga,et al.  Dynamic Control of Response Criterion in Premotor Cortex during Perceptual Detection under Temporal Uncertainty , 2015, Neuron.

[14]  Thomas Parr,et al.  Transferring structural knowledge across cognitive maps in humans and models , 2019, Nature Communications.

[15]  Jin Wang,et al.  Potential landscape and flux framework of nonequilibrium networks: Robustness, dissipation, and coherence of biochemical oscillations , 2008, Proceedings of the National Academy of Sciences.

[16]  Alex Proekt,et al.  Universality of macroscopic neuronal dynamics in Caenorhabditis elegans , 2017 .

[17]  Laurie D. Burns,et al.  High-speed, miniaturized fluorescence microscopy in freely moving mice , 2008, Nature Methods.

[18]  Surya Ganguli,et al.  On simplicity and complexity in the brave new world of large-scale neuroscience , 2015, Current Opinion in Neurobiology.

[19]  Chethan Pandarinath,et al.  Neural population dynamics in human motor cortex during movements in people with ALS , 2015, eLife.

[20]  Naoshige Uchida,et al.  Demixed principal component analysis of neural population data , 2014, eLife.

[21]  Ronen Talmon,et al.  Empirical intrinsic geometry for nonlinear modeling and time series filtering , 2013, Proceedings of the National Academy of Sciences.

[22]  Omri Barak,et al.  Recurrent neural networks as versatile tools of neuroscience research , 2017, Current Opinion in Neurobiology.

[23]  Ronen Talmon,et al.  PNAS Plus Significance Statements , 2017, Proceedings of the National Academy of Sciences.

[24]  Xue-Xin Wei,et al.  Emergence of grid-like representations by training recurrent neural networks to perform spatial localization , 2018, ICLR.

[25]  Razvan Pascanu,et al.  Vector-based navigation using grid-like representations in artificial agents , 2018, Nature.

[26]  Bernhard Schölkopf,et al.  Nonlinear Component Analysis as a Kernel Eigenvalue Problem , 1998, Neural Computation.

[27]  R. Romo,et al.  Neuronal correlates of parametric working memory in the prefrontal cortex , 1999, Nature.

[28]  Surya Ganguli,et al.  Discovering Precise Temporal Patterns in Large-Scale Neural Recordings through Robust and Interpretable Time Warping , 2019, Neuron.

[29]  Matthew T. Kaufman,et al.  Neural population dynamics during reaching , 2012, Nature.

[30]  Christopher D. Harvey,et al.  Choice-specific sequences in parietal cortex during a virtual-navigation decision task , 2012, Nature.

[31]  J. DiCarlo,et al.  Using goal-driven deep learning models to understand sensory cortex , 2016, Nature Neuroscience.

[32]  Alan L. Yuille,et al.  Non-negative matrix factorization of multimodal MRI, fMRI and phenotypic data reveals differential changes in default mode subnetworks in ADHD , 2014, NeuroImage.

[33]  V. Jayaraman,et al.  Intensity versus Identity Coding in an Olfactory System , 2003, Neuron.

[34]  J. Tenenbaum,et al.  A global geometric framework for nonlinear dimensionality reduction. , 2000, Science.

[35]  Drew N. Robson,et al.  Brain-wide neuronal dynamics during motor adaptation in zebrafish , 2012, Nature.

[36]  Demis Hassabis,et al.  Improved protein structure prediction using potentials from deep learning , 2020, Nature.

[37]  Benjamin F. Grewe,et al.  High-speed in vivo calcium imaging reveals neuronal network activity with near-millisecond precision , 2010, Nature Methods.

[38]  W. Newsome,et al.  Context-dependent computation by recurrent dynamics in prefrontal cortex , 2013, Nature.

[39]  Matthew T. Kaufman,et al.  A neural network that finds a naturalistic solution for the production of muscle activity , 2015, Nature Neuroscience.

[40]  Michael J. Berry,et al.  Mapping a Complete Neural Population in the Retina , 2012, The Journal of Neuroscience.

[41]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[42]  Matthew T. Kaufman,et al.  Supplementary materials for : Cortical activity in the null space : permitting preparation without movement , 2014 .

[43]  Christopher D. Harvey,et al.  Recurrent Network Models of Sequence Generation and Memory , 2016, Neuron.

[44]  Byron M. Yu,et al.  Neural constraints on learning , 2014, Nature.

[45]  A. Cheng,et al.  simultaneous two-photon calcium imaging at different depths with spatiotemporal multiplexing , 2011 .

[46]  H. Sompolinsky,et al.  Chaos in Neuronal Networks with Balanced Excitatory and Inhibitory Activity , 1996, Science.

[47]  Surya Ganguli,et al.  Universality and individuality in neural dynamics across large populations of recurrent networks , 2019, NeurIPS.

[48]  T. Holy,et al.  Fast Three-Dimensional Fluorescence Imaging of Activity in Neural Populations by Objective-Coupled Planar Illumination Microscopy , 2008, Neuron.

[49]  Wojciech M. Czarnecki,et al.  Grandmaster level in StarCraft II using multi-agent reinforcement learning , 2019, Nature.

[50]  Chethan Pandarinath,et al.  Inferring single-trial neural population dynamics using sequential auto-encoders , 2017, Nature Methods.

[51]  B. Nadler,et al.  Diffusion maps, spectral clustering and reaction coordinates of dynamical systems , 2005, math/0503445.

[52]  T. Sideris Ordinary Differential Equations and Dynamical Systems , 2013 .

[53]  Michael J. Berry,et al.  The simplest maximum entropy model for collective behavior in a neural network , 2012, 1207.6319.

[54]  Connor Brennan,et al.  A quantitative model of conserved macroscopic dynamics predicts future motor commands , 2019, eLife.

[55]  Stéphane Lafon,et al.  Diffusion maps , 2006 .

[56]  Michael J. Berry,et al.  Weak pairwise correlations imply strongly correlated network states in a neural population , 2005, Nature.

[57]  Annika L A Nichols,et al.  A global brain state underlies C. elegans sleep behavior , 2017, Science.

[58]  Philipp J. Keller,et al.  Whole-brain functional imaging at cellular resolution using light-sheet microscopy , 2013, Nature Methods.

[59]  Edsger W. Dijkstra,et al.  A note on two problems in connexion with graphs , 1959, Numerische Mathematik.

[60]  Geoffrey E. Hinton,et al.  Similarity of Neural Network Representations Revisited , 2019, ICML.

[61]  Matteo Marsili,et al.  The Stochastic Complexity of Spin Models: Are Pairwise Models Really Simple? , 2017, Entropy.