Moving beyond generalization to accurate interpretation of flexible models

Genkin, Mikhail; Engel, Tatiana A.

doi:10.1038/s42256-020-00242-6

Cited by 21 publications

(31 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The validation of inferred circuit mechanisms via RNN perturbations is critical, because the fit quality alone does not guarantee that the inferred model captures the correct mechanism that generated data 32 . Moreover, it is generally uncertain whether interpretable circuit mechanisms exist in RNNs optimized to perform cognitive tasks.…”

Section: Resultsmentioning

confidence: 99%

Latent circuit inference from heterogeneous neural responses during cognitive tasks

Langdon

Engel

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Higher cortical areas carry a wide range of sensory, cognitive, and motor signals supporting complex goal-directed behavior. These signals are mixed in heterogeneous responses of single neurons tuned to multiple task variables. Dimensionality reduction methods used to analyze neural responses rely merely on correlations, leaving unknown how heterogeneous neural activity arises from connectivity to drive behavior. Here we present a framework for inferring a low-dimensional connectivity structure—the latent circuit—from high-dimensional neural response data. The latent circuit captures mechanistic interactions between task variables and their mixed representations in single neurons. We apply the latent circuit inference to recurrent neural networks trained to perform a context-dependent decision-making task and find a suppression mechanism in which contextual representations inhibit irrelevant sensory responses. We validate this mechanism by confirming the behavioral effects of patterned connectivity perturbations predicted by the latent circuit structure. Our approach can reveal interpretable and causally testable circuit mechanisms from heterogeneous neural responses during cognitive tasks.

show abstract

Section: Resultsmentioning

confidence: 99%

Latent circuit inference from heterogeneous neural responses during cognitive tasks

Langdon

Engel

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The likelihood L [Y (t)|θ] is a conditional probability of observing the data Y (t) given a model θ = {Φ(x), p 0 (x), D}. The likelihood is obtained by marginalizing the joint probability density P (X (t), Y (t)|θ) over all possible latent trajectories X (t) that may underlie the data 2,20 :…”

Section: Methodsmentioning

confidence: 99%

“…Such systems are commonly described by Langevin dynamics, in which deterministic forces define persistent collective trends and noise captures fast microscopic interactions 1 . Langevin equations are used to model stochastic evolution of complex systems such as neural networks [2][3][4][5] , motile cells 6 , swarming animals 7 , carbon nanotubes 8 , financial markets 9 , or climate dynamics 10 . While such systems can be readily observed in experiments or microscopic simulations, the analytical form of the Langevin equation usually cannot be easily derived from microscopic models or physical principles.…”

mentioning

confidence: 99%

“…However, in many complex systems, the trajectories cannot be observed directly, but are only gleaned from a stochastic observation process that depends on the latent Langevin dynamics. For example, spikes recorded from neurons in the brain form stochastic point processes with statistics controlled by the collective dynamics of the surrounding network 2,16,17 . Similarly, the dynamics of a protein are observed through photons emitted by fluorescent dyes tagging the protein in single-molecule microscopy experiments [18][19][20] .…”

mentioning

confidence: 99%

“…The inference with latent dynamical models is data efficient as it integrates information along the entire latent trajectory, but it may be sensitive to the data distribution. Previous work only considered the inference of latent Langevin dynamics for equilibrium systems with the steadystate data distribution 2,20 . Whether these methods extend to non-equilibrium systems has not been tested.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Learning non-stationary Langevin dynamics from stochastic observations of latent trajectories

Genkin,

Hughes,

Engel

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Many complex systems operating far from the equilibrium exhibit stochastic dynamics that can be described by a Langevin equation. Inferring Langevin equations from data can reveal how transient dynamics of such systems give rise to their function. However, dynamics are often inaccessible directly and can be only gleaned through a stochastic observation process, which makes the inference challenging. Here we present a nonparametric framework for inferring the Langevin equation, which explicitly models the stochastic observation process and non-stationary latent dynamics. The framework accounts for the non-equilibrium initial and final states of the observed system and for the possibility that the system's dynamics define the duration of observations. Omitting any of these non-stationary components results in incorrect inference, in which erroneous features arise in the dynamics due to non-stationary data distribution. We illustrate the framework using models of neural dynamics underlying decision making in the brain.

show abstract

A Marr's Three‐Level Analytical Framework for Neuromorphic Electronic Systems

Guo

Zou

et al. 2021

Advanced Intelligent Systems

View full text Add to dashboard Cite

Neuromorphic electronics, an emerging field that aims for building electronic mimics of the biological brain, holds promise for reshaping the frontiers of information technology and enabling a more intelligent and efficient computing paradigm. As their biological brain counterpart, the neuromorphic electronic systems are complex, having multiple levels of organization. Inspired by David Marr's famous three‐level analytical framework developed for neuroscience, the advances in neuromorphic electronic systems are selectively surveyed and given significance to these research endeavors as appropriate from the computational level, algorithmic level, or implementation level. Under this framework, the problem of how to build a neuromorphic electronic system is defined in a tractable way. In conclusion, the development of neuromorphic electronic systems confronts a similar challenge to the one neuroscience confronts, that is, the limited constructability of the low‐level knowledge (implementations and algorithms) to achieve high‐level brain‐like (human‐level) computational functions. An opportunity arises from the communication among different levels and their codesign. Neuroscience lab‐on‐neuromorphic chip platforms offer additional opportunity for mutual benefit between the two disciplines.

show abstract

Moving beyond generalization to accurate interpretation of flexible models

Cited by 21 publications

References 42 publications

Latent circuit inference from heterogeneous neural responses during cognitive tasks

Latent circuit inference from heterogeneous neural responses during cognitive tasks

Learning non-stationary Langevin dynamics from stochastic observations of latent trajectories

A Marr's Three‐Level Analytical Framework for Neuromorphic Electronic Systems

Contact Info

Product

Resources

About