On simplicity and complexity in the brave new world of large-scale neuroscience.

By Peiran Gao, Surya Ganguli. 2015

408 citations recorded by the source corpus.

Technological advances have dramatically expanded our ability to probe multi-neuronal dynamics and connectivity in the brain. However, our ability to extract a simple conceptual understanding from complex data is increasingly hampered by the lack of theoretically principled data analytic procedures, as well as theoretical frameworks for how circuit connectivity and dynamics can conspire to generate emergent behavioral and cognitive functions. We review and outline potential avenues for progress, including new theories of high dimensional data analysis, the need to analyze complex artificial networks, and methods for analyzing entire spaces of circuit models, rather than one model at a time. Such interplay between experiments, data analysis and theory will be indispensable in catalyzing conceptual advances in the age of large-scale neuroscience.

Subfields: Cognitive Neuroscience, Electrical and Electronic Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Electrical and Electronic Engineering

Citations in the corpus

Orthogonal representations for robust context-dependent task performance in brains and neural networks (2022, neutral): “However, how the neural representations that support context-dependent task performance are acquired remains a key open question for cognitive and neural scientists (Gao and Ganguli, 2015; Mastrogiuseppe andOstojic, 2018; Cueva et al., 2020; Dubreuil et al., 2020; Badre et al., 2021; Freund et al., ”
Re-evaluating circuit mechanisms underlying pattern separation (2019, neutral): “However, subsampling of neuronal populations and limited recording times can lead to severe underestimates (Gao and Ganguli, 2015), while measurement noise and nonlinear dynamics can inflate the apparent dimensionality (Cunningham and Yu, 2016).”
Low-dimensional dynamics for working memory and time encoding (2020, background): “67–69) or as high-dimensional as it could be (70).”
Unsupervised Discovery of Demixed, Low-Dimensional Neural Dynamics across Multiple Timescales through Tensor Component Analysis (2017, methodology): “Neuroscientists have often turned to unbiased dimensionality reduction methods to understand these complex 27 datasets [12, 13].”
Why neurons mix: high dimensionality for higher cognition. (2016, background): “The formalism of [21] can be extended to continuous variables that are”
Linking Connectivity, Dynamics, and Computations in Low-Rank Recurrent Neural Networks. (2017, neutral): “Even if one were able to experimentally record the activity of every neuron and the strength of each synapse in a behaving animal, understanding the causal relationships between these quantities would remain a daunting challenge because an appropriate conceptual framework is currently lacking (Gao a”
Shared cortex-cerebellum dynamics in the execution and learning of a motor task (2019, neutral): “A direction for future study would be to determine how L5GrC correlations and dimensionality depend on task complexity (Gao and Ganguli, 2015).”
Interpreting neural computations by examining intrinsic and embedding dimensionality of neural activity (2021, background): “This mismatch makes signals across neurons highly redundant and raises the question of how the population activity represents task-relevant variables [6–8].”
Training Excitatory-Inhibitory Recurrent Neural Networks for Cognitive Tasks: A Simple and Flexible Framework (2016, background): “At the same time, the novel challenges posed by high-dimensional neural data have led to the development of new methods for analyzing and modeling such data [10 – 12].”
A connectome of the Drosophila central complex reveals network motifs suitable for flexible navigation and context-dependent action selection (2020, background): “…learning, a broader question concerns the level of understanding that is appropriate or even possible for the function of large and complex neural networks (Gao and Ganguli, 2015; Hasson et al., 2020; Lillicrap and Kording, 2019; Richards et al., 2019; Saxe et al., 2020; Yamins and DiCarlo, 2016).”