It will be shown that according to theorems of K. Menger, every neuron grid if identified with a curve is able to preserve the adopted qualitative structure of a data space. Furthermore, if this identification is made, the neuron grid structure can always be mapped to a subset of a universal neuron grid which is constr…
SeReNe prunes neurons with low sensitivity to reduce network size.
problem Large neural networks consume too many resources on resource-constrained devices.
method Exploits neural sensitivity as a regularizer to prune neurons with low sensitivity.
result Pruning neurons with low sensitivity achieves competitive compression ratios.
In the wake of recent advances in experimental methods in neuroscience, the ability to record in-vivo neuronal activity from awake animals has become feasible. The availability of such rich and detailed physiological measurements calls for the development of advanced data analysis tools, as commonly used techniques do …
Neural circuits contain heterogeneous groups of neurons that differ in type, location, connectivity, and basic response properties. However, traditional methods for dimensionality reduction and clustering are ill-suited to recovering the structure underlying the organization of neural circuits. In particular, they do n…
Quadratic neurons enhance deep networks' approximation capabilities.
problem Improving deep networks' expressive power and efficiency.
method Introduced quadratic neurons and analyzed their performance in deep quadratic networks.
result Quadratic networks can approximate functions more efficiently and uniquely than conventional networks.
Morphological neurons are powerful and versatile for classification and regression problems.
problem Designing effective sequences of morphological operations and structuring elements.
method Theoretical analysis of morphological neurons as a sum of hinge functions, and their effectiveness in approximating any continuous function.
result Morphological neurons are more powerful than previously anticipated and can approximate any continuous function.
Simplified neural network EFTs reveal a single critical condition.
problem Understanding neuron statistics in neural networks at initialization.
method Diagrammatic approach to effective field theories (EFTs).
result A single condition governs criticality of all neuron preactivations.
RNNs trained on head direction task mimic brain's compass and shifter neurons.
problem Modeling brain's head direction system using neural networks.
method Optimized recurrent neural networks trained on angular velocity integration.
result RNNs naturally emerge with compass and shifter neuron-like properties.
We sparsify gated RNNs by simplifying their structure.
problem Improving efficiency of RNNs by reducing their complexity.
method Adjust existing sparsification techniques to gated RNNs, sparsifying preactivations of gates.
result Simplified LSTM structure improves model performance and efficiency.
Quadratic autoencoder improves low-dose CT image denoising.
problem Low-dose CT image denoising.
method Quadratic autoencoder architecture applied to CT denoising.
result Quadratic autoencoder achieves superior denoising performance and efficiency.
Under-parameterized networks can either copy or average teacher weights, leading to universal optimal solutions.
problem Approximating a teacher network with an under-parameterized student network.
method Analyzing shallow neural networks with erf activation function and unitary teacher weights, proving copy-average configurations are critical points and finding the optimal solution.
result The optimal solution for under-parameterized networks has a universal structure, whether copying or averaging teacher neurons.
Estimates neuronal connectivity from spike times using flexible Hawkes processes.
problem Learning latent network structure from multivariate point process data.
method Proposes a new nonstationary Hawkes process and uses sparse least squares estimation.
result Establishes non-asymptotic error bounds and selection consistency for estimated parameters.
A new method for growing neural networks by splitting neurons, improving efficiency.
problem Optimizing neural network structures, especially for lightweight architectures.
method A progressive training approach using steepest descent to adaptively grow and split neurons.
result The method provides a computationally efficient way to optimize neural network structures.
Unified approach for neural networks with multi-compartmental neurons and non-Hebbian plasticity.
problem Limited computational power of existing neural network models for multi-compartmental neurons and non-Hebbian plasticity.
method Unified extension of similarity matching approach to derive neural networks with multi-compartmental neurons and local, non-Hebbian learning rules.
result Unified approach facilitates understanding of multi-compartmental neuronal structures and non-Hebbian plasticity.
This paper compares spike sorting techniques for rat brain neuronal activity.
problem Improving the accuracy of spike sorting for neuronal activity analysis.
method Three-step spike sorting process: detection, feature extraction, and clustering. Various methods are compared.
result Kernel PCA outperforms in feature extraction, leading to better spike sorting results.
Structured pruning method reduces RNN sizes and speeds up inference.
problem Large RNN models are hard to deploy on edge devices.
method Structured pruning through neuron selection, minimizing L0 norm of weight matrix.
result Nearly 20x speedup achieved without performance loss.
Graph Spectral Regularization makes neural network layers more interpretable.
problem Making neural network layers more interpretable without sacrificing performance.
method Using a graph Laplacian penalty to structure hidden layer activations.
result Encourages smooth activations within hidden layers, leading to better interpretability.
Automatically designs neural network structure using matrix conditioning.
problem Choosing an effective size and structure of neural networks for new datasets is time-consuming.
method Adjusts neuron proportions and scales network size using matrix conditioning.
result Small networks achieve high accuracy on various datasets.
Neural connectomics has begun producing massive amounts of data, necessitating new analysis methods to discover the biological and computational structure. It has long been assumed that discovering neuron types and their relation to microcircuitry is crucial to understanding neural function. Here we developed a nonpara…
New method clusters neurons with similar connectivity profiles.
problem Accurately determining which neurons have similar neurological tasks.
method Proposes clustered Gaussian graphical model and symmetric convex clustering penalty.
result Demonstrates effectiveness of the approach on synthetic and real-world data.
New artificial neuron doubles weight for improved deep learning accuracy.
problem Improving deep learning models' accuracy.
method Introducing a double-weight neuron, tested on MNIST and CIFAR-10 datasets.
result Significant improvement in classification accuracy for MNIST and CIFAR-10 datasets.
Model detects patterns in noisy binary data, explaining neuron activity in terms of cell assemblies.
problem Detecting structure in noisy or approximate repeats of patterns in sparse binary data.
method Probabilistic binary latent variable model based on Noisy-OR model, inferring sparse activity in latent variables.
result Model successfully extracts and explains latent structure in spiking neural data.
STNMF method uncovers neural circuit components in retinal ganglion cells.
problem Deciphering complex neuronal circuit components in the brain.
method Spike-triggered non-negative matrix factorization (STNMF) method.
result STNMF can detect various properties of upstream bipolar cells and recover synaptic connection strengths.
Persistent neurons improve neural network optimization by leveraging previous solutions.
problem Improving neural network optimization under different initialization and data distributions.
method Persistent neurons use information from previous converged solutions to explore new landscapes and avoid local minima.
result Persistent neurons converge to more optimal solutions and improve model performance under various initializations.
New method makes neural networks transparent, revealing learning modes.
problem Lack of interpretability in neural networks.
method Weight pathway analysis (WPA) to decompose neural networks into subnetworks.
result Neural networks store and utilize information holographically, with linear and nonlinear learning modes.
Paper introduces supervised and unsupervised TAM models for binary neurons.
problem Learning and retrieval of structured triplets of patterns in neural networks.
method Extends Hebbian paradigm to supervised and unsupervised protocols, using glassy statistical mechanical techniques.
result Obtained self-consistency equations for critical dataset sizes and retrieval performance.
Gradient descent slows significantly in over-parameterized single neuron learning.
problem Learning a single neuron with over-parameterization and square loss.
method Analysis of gradient descent dynamics, proving convergence rates and lower bounds.
result Over-parameterization can exponentially slow down the convergence rate of gradient descent.
A measure of neural complexity quantifies how hard it is to access information across neurons.
problem Understanding how mutual information is distributed among neurons in neural networks.
method Partial Information Decomposition (PID) to disentangle contributions of single neurons, multiple neurons, and synergistic effects.
result Representational Complexity measures the difficulty of accessing information across multiple neurons.
Complex neural networks simplify to a mean field model as the number of neurons increases.
problem Understanding the behavior of multilayer neural networks with many neurons.
method Developed a mean field limit formalism for multilayer neural networks under stochastic gradient descent.
result The behavior of multilayer neural networks simplifies to a mean field model as the number of neurons grows large.
TENP prunes experts and neurons in Mixture-of-Experts models for efficient deployment.
problem Efficient deployment of large language models constrained by static parameter footprint.
method Structured Trapezoidal ExpertNeuron Pruning (TENP) identifies and retains important experts and neurons.
result DeepSeek model achieves 10% better performance on code generation tasks with 40% expert sparsity.
With the success of deep learning, recent efforts have been focused on analyzing how learned networks make their classifications. We are interested in analyzing the network output based on the network structure and information flow through the network layers. We contribute an algorithm for 1) analyzing a deep network t…
A new algorithm trains deep neural networks by adding neurons greedily.
problem Training deep neural networks efficiently and effectively.
method Neuron Pursuit (NP) algorithm, which alternates between neuron addition and loss minimization.
result The algorithm can train deep neural networks efficiently and effectively.
Neuron Shapley identifies key neurons in deep networks, improving model accuracy and fairness.
problem Identifying responsible neurons in deep networks for better model performance and fairness.
method Neuron Shapley framework quantifies neuron contributions, accounting for interactions.
result Removing just 30 critical filters can destroy model accuracy, revealing network function.
Deep convolutional neural networks (CNNs) have demonstrated impressive performance on visual object classification tasks. In addition, it is a useful model for predication of neuronal responses recorded in visual system. However, there is still no clear understanding of what CNNs learn in terms of visual neuronal circu…
New method decomposes sensory information from neurons into specific stimuli and features.
problem Understanding how much and what specific information neurons encode.
method Introduced axioms for meaningful stimulus-wise decomposition and derived a tractable solution using diffusion models.
result Can efficiently estimate contributions of specific stimuli and features to encoded information.
Describes explaining neurons in deep representations using compositional logical concepts.
problem Interpreting neuron behavior in deep neural networks.
method Identifying compositional logical concepts that closely approximate neuron behavior.
result Compositional explanations provide insights into model performance and allow for adversarial example creation.
Deep P-Spline automates DNN structure selection for complex regression problems.
problem Challenges in selecting optimal network structures for DNNs.
method Linking neuron selection to knot placement in basis expansion techniques, introducing a difference penalty for automated knot selection.
result Deep P-Spline extends model class and forms a latent variable modeling framework with theoretical guarantees.
DEUs learn nonlinear activation functions from data, reducing network size.
problem Fixed activation functions in neural networks limit performance.
method Differential equation units (DEUs) learn nonlinear activation functions from data.
result DEUs enable neurons to change their activation functions during training.
New neural network uses differential equations for adaptable activation functions.
problem Fixed activation functions limit neural network performance and size.
method Introduces differential equation units (DEUs) that learn nonlinear activation functions.
result DEUs enable more compact networks with comparable performance.
Dynamics and function of neuronal networks are determined by their synaptic connectivity. Current experimental methods to analyze synaptic network structure on the cellular level, however, cover only small fractions of functional neuronal circuits, typically without a simultaneous record of neuronal spiking activity. H…
This research investigates selectively pruning hyper and hypo neurons to improve neural network generalization.
problem Improving neural network generalization to unseen data.
method Investigates pruning hyper and hypo neurons selectively in fully connected layers of CNNs.
result Selective pruning of hyper and hypo neurons improves model performance on out-of-domain data.
Modeling hidden neurons in SNNs using mesoscopic approximations.
problem Underconstrained problem of modeling unobserved neurons in SNNs.
method Coarse-graining and mean-field approximations to derive neuLVM.
result neuLVM can efficiently model large SNNs and recover connectivity parameters.
Visualizes deep neural networks for speech recognition using learned topographic filter maps.
problem Unintuitive internal structure of deep neural networks complicates activation visualization.
method Trains a convolutional speech recognition model with filters arranged in a 2D grid, highlighting similar filters.
result Topographic filter maps visualize artificial neuron activations more intuitively.
Topological methods improve neuron analysis and tracer injection summary.
problem Traditional methods fail to capture the tree-like structure of neurons.
method Discrete Morse (DM) Theory for neuron skeletonization and consensus tree summarization.
result Significant performance improvements over non-topological methods.
Bayesian method sparsifies gated RNNs, improving speed and interpretability.
problem Sparsifying neural networks to reduce complexity and improve performance.
method Bayesian approach to sparsify weights, neurons, and gates in LSTM architectures.
result Sparsified gated RNNs speed up forward pass and improve compression.
We propose a new generic type of stochastic neurons, called q-neurons, that considers activation functions based on Jackson's q-derivatives with stochastic parameters q. Our generalization of neural network architectures with q-neurons is shown to be both scalable and very easy to implement. We demonstrate expe…
We developed a faster method for calculating neuron importance in neural networks.
problem Assigning importance to individual neurons in deep learning models.
method We developed Neuron Integrated Gradients, a scalable implementation of Total Conductance.
result Neuron Integrated Gradients is faster and empirically stronger than DeepLIFT.
This paper presents a Bayesian approach to learning the connectivity structure of a group of neurons from data on configuration frequencies. A major objective of the research is to provide statistical tools for detecting changes in firing patterns with changing stimuli. Our framework is not restricted to the well-under…