We present an open-source tool for visualizing multi-head self-attention in Transformer-based language representation models. The tool extends earlier work by visualizing attention at three levels of granularity: the attention-head level, the model level, and the neuron level. We describe how each of these views can he…
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,742 papers · 148 categories
Trend · papers per month
48 results for “Neuronal Attention Model”
New NAM model improves image processing speed and accuracy.
problem Time-consuming pixel-by-pixel analysis of large images.
method Integrates Neuronal Attention model (NAM) to focus on small regions.
result Efficiently processes large images with improved speed and accuracy.
GraphDINO learns neuronal morphologies from unlabeled data.
problem Unsupervised learning of neuronal morphologies from unlabeled data.
method Transformer-based approach with novel attention mechanism and data augmentation.
result GraphDINO yields morphological clusterings on par with expert classification.
This work explains the structural origins of attention sinks in LLMs.
problem Initial tokens disproportionately monopolize attention scores in LLMs.
method Traced to self-attention's value aggregation process and FFN layer activations.
result Attention sinks form due to variance discrepancy and dimension disparity.
Explains deep learning models and their geometric properties.
problem Understanding the geometric intuition behind deep learning models.
method Geometrical intuition and novel insights into loss surfaces of deep learning models.
result Deep neural networks carve out manifolds with multiplication neurons.
Spark Transformer achieves high sparsity in FFN and attention without sacrificing model quality.
problem Inactivity of most neurons in FFN of Transformers, leading to inefficiency.
method Top-k masking, statistical top-k algorithm, and predictor for activated entries.
result Only 8% of FFN neurons are activated, achieving up to 2.5x wall-time speedup.
The predictive power of neural networks often costs model interpretability. Several techniques have been developed for explaining model outputs in terms of input features; however, it is difficult to translate such interpretations into actionable insight. Here, we propose a framework to analyze predictions in terms of …
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…
Improved Transformer performance by addressing 'explaining away' effect.
problem Transformer's self-attention mechanism can explain away important input features.
method Proposed a doubly-normalized attention scheme to avoid 'explaining away' effect.
result Improved performance on benchmarks with the new attention scheme.
Network pruning is aimed at imposing sparsity in a neural network architecture by increasing the portion of zero-valued weights for reducing its size regarding energy-efficiency consideration and increasing evaluation speed. In most of the conducted research efforts, the sparsity is enforced for network pruning without…
DCNs mimic neuronal networks for improved neural classification.
problem Lack of topological similarity between DNNs and biological neural networks.
method Developed DCNs with topologies inspired by real-world neuronal networks.
result High classification accuracy achieved by DCNs.
Stable unactivated neurons reduce expressiveness in ReLU networks.
problem Reducing expressiveness in ReLU neural networks due to stably unactivated neurons.
method Investigated the probability of neurons being stably unactivated in ReLU networks with symmetric weight and bias distributions.
result Proved the probability of a neuron being stably unactivated in the second hidden layer of a ReLU network.
This paper investigates the effect of leak in spiking neural networks.
problem Comparative analysis of leaky and non-leaky spiking neuron models.
method Experimental and frequency domain analysis of spiking neural networks.
result Leaky spiking neuron models provide improved robustness and generalization but decrease sparsity of computation.
AKOrN uses synchronized neurons to improve AI tasks.
problem Improving AI performance through better neural representations.
method AKOrN introduces synchronized neurons to replace threshold units.
result AKOrN improves performance across various AI tasks.
Regularity Normalization: Neuroscience-Inspired Unsupervised Attention across Neural Network Layerscs.LG
Inspired by the adaptation phenomenon of neuronal firing, we propose the regularity normalization (RN) as an unsupervised attention mechanism (UAM) which computes the statistical regularity in the implicit space of neural networks under the Minimum Description Length (MDL) principle. Treating the neural network optimiz…
Perceptual capabilities of artificial systems have come a long way since the advent of deep learning. These methods have proven to be effective, however they are not as efficient as their biological counterparts. Visual attention is a set of mechanisms that are employed in biological visual systems to ease computationa…
BDH model learns like the brain, rivaling Transformer performance.
problem Leveraging brain-like properties for machine learning.
method Scale-free biologically inspired network of neuron particles.
result BDH model achieves Transformer-like performance with interpretability.
Capsule networks improve performance on image classification tasks with fewer parameters.
problem Improving performance of capsule networks with fewer parameters.
method Inverted dot-product attention routing, Layer Normalization, concurrent iterative routing.
result Improves performance on benchmark datasets CIFAR-10 and CIFAR-100, and performs at-par with ResNet-18.
This study approximates neural network features for modeling relations and attention mechanisms.
problem Approximating neural network features for modeling relations and attention mechanisms.
method Analyzes inner products of multi-layer perceptrons for universal approximation of symmetric and asymmetric relation functions.
result Universal approximation of relation functions and attention mechanisms using inner products of neural networks.
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.
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.
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.
Paper analyzes infinite-width attention layers using Tensor Programs.
problem Capturing the infinite-width limit of attention layers.
method Tensor Programs framework to rigorously identify the limit distribution.
result Derives exact form of infinite-width limit distribution without Gaussian approximations.
AGF explains feature learning in neural networks through alternating steps.
problem Understanding what features neural networks learn and how they learn them.
method AGF is an algorithmic framework that approximates the dynamics of feature learning in two-layer networks.
result AGF provides a unified framework to understand feature learning in neural networks, matching experimental results across various architectures.
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.
Neural responses are highly variable, and some portion of this variability arises from fluctuations in modulatory factors that alter their gain, such as adaptation, attention, arousal, expected or actual reward, emotion, and local metabolic resource availability. Regardless of their origin, fluctuations in these signal…
Model improves information transfer from visual streams.
problem Challenges in unsupervised learning from continuous visual data.
method Inspired by physics, maximizes mutual information through temporal process.
result Focus of attention enhances information transfer from input stream.
Deep neural networks (DNNs) are known for extracting useful information from large amounts of data. However, the representations learned in DNNs are typically hard to interpret, especially in dense layers. One crucial issue of the classical DNN model such as multilayer perceptron (MLP) is that neurons in the same layer…
Inspired by complexity and diversity of biological neurons, our group proposed quadratic neurons by replacing the inner product in current artificial neurons with a quadratic operation on input data, thereby enhancing the capability of an individual neuron. Along this direction, we are motivated to evaluate the power o…
This paper improves model fusion by training-time neuron alignment, reducing barriers in multi-model fusion.
problem Diverse neuron permutations across different settings hinder model fusion performances.
method Training-time neuron alignment using fixed neuron anchors to reduce training-time permutations.
result Training-time neuron alignment improves fusion of pretrained models and federated learning performances.
To understand how rich dynamics emerge in neural populations, we require models exhibiting a wide range of activity patterns while remaining interpretable in terms of connectivity and single-neuron dynamics. However, it has been challenging to fit such mechanistic spiking networks at the single neuron scale to empirica…
Single neuron sensitivity reveals model weaknesses.
problem Understanding model susceptibility to adversarial attacks.
method Analyzing sensitivity of individual neurons to perturbations.
result Single neuron attacks are as effective as full model attacks.
Novel 'strong neuron' improves deep learning efficiency and robustness.
problem Improving deep learning efficiency and robustness against adversarial attacks.
method Introducing a novel 'strong neuron' model and a constructive training algorithm.
result Achieved 10x-100x reduction in operations count and hardware requirements.
This paper presents a new artificial neuron model capable of learning its receptive field in the topological domain of inputs. The model provides adaptive and differentiable local connectivity (plasticity) applicable to any domain. It requires no other tool than the backpropagation algorithm to learn its parameters whi…
Single neuron with ADA learns XOR and outperforms other functions.
problem Classifying linearly non-separable data.
method Proposed a new artificial neuron with apical dendrite activation.
result ADA function achieves 100% accuracy on XOR and superior performance on benchmark datasets.
Single-spike neurons can approximate as well as multi-spike neurons.
problem Limitation of single-spike neurons in spiking neural networks.
method Comparison of single-spike and multi-spike neural networks.
result Single-spike and multi-spike neural networks are equivalent in approximation capabilities.
Panda predicts chaotic systems without retraining, showing emergent properties.
problem Predicting chaotic systems with small errors.
method Trained on a synthetic dataset of chaotic dynamical systems using evolutionary algorithms.
result Panda predicts unseen chaotic systems with zero-shot learning.
Improved Bayesian inference for neuronal ensemble inference reduces computational cost.
problem Efficient inference of neuronal ensembles from activity data.
method Modified MCMC algorithm with simulated annealing for hyperparameter control.
result Our method reduces computational cost while maintaining or improving inference accuracy.
Neurons in cortical circuits exhibit coordinated spiking activity, and can produce correlated synchronous spikes during behavior and cognition. We recently developed a method for estimating the dynamics of correlated ensemble activity by combining a model of simultaneous neuronal interactions (e.g., a spin-glass model)…
Neural networks learn to mimic brain neurons with two-input activation functions, improving performance and robustness.
problem Training neural networks to mimic the complex interactions of brain neurons.
method Developed a network-in-network architecture with two-input activation functions, optimized hyperparameters, and compared to conventional ReLU networks.
result Two-input activation functions can learn soft XOR functions, improving network performance and robustness.
Novel method decorrelates neurons for better deep learning model generalization.
problem High correlations between neurons limit deep learning model generalization.
method Regularization terms from minimum spanning tree of neuron cliques, using correlation dissimilarities.
result Our regularizers outperform existing methods and minimize neuron redundancies.
The artificial neural network is a popular framework in machine learning. To empower individual neurons, we recently suggested that the current type of neurons could be upgraded to 2nd order counterparts, in which the linear operation between inputs to a neuron and the associated weights is replaced with a nonlinear qu…
CHANI learns classification tasks with local transformations inspired by biology.
problem Proving neural networks can learn classification tasks with local transformations.
method CHANI uses spiking neurons modeled by Hawkes processes with expert aggregation for local learning.
result CHANI can learn and encode multiple classes, forming assemblies of neurons.
Persistent Neuronscs.LG
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.
We propose the BinaryGAN, a novel generative adversarial network (GAN) that uses binary neurons at the output layer of the generator. We employ the sigmoid-adjusted straight-through estimators to estimate the gradients for the binary neurons and train the whole network by end-to-end backpropogation. The proposed model …
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…
The challenge of assigning importance to individual neurons in a network is of interest when interpreting deep learning models. In recent work, Dhamdhere et al. proposed Total Conductance, a "natural refinement of Integrated Gradients" for attributing importance to internal neurons. Unfortunately, the authors found tha…
Researchers develop methods to learn neuron dynamics from colored noise.
problem Learning nonlocal stochastic neuron dynamics from colored noise.
method Proposed two methods for closing Fokker-Planck equations: nonlocal large-eddy-diffusivity closure and data-driven sparse regression.
result Mutual information and total correlation between stimulus and neuron states calculated for FHN neuron.