Results 41 to 50 of about 65,893 (254)

MAMGD: Gradient-Based Optimization Method Using Exponential Decay

Technologies
Optimization methods, namely, gradient optimization methods, are a key part of neural network training. In this paper, we propose a new gradient optimization method using exponential decay and the adaptive learning rate using a discrete second-order ...
Nikita Sakovich, Dmitry Aksenov, Ekaterina Pleshakova, Sergey Gataullin   +3 more
doaj   +1 more source

Application of machine learning to viscoplastic flow modeling

, 2018
We present a method to construct reduced-order models for duct flows of Bingham media. Our method is based on proper orthogonal decomposition (POD) to find a low-dimensional approximation to the velocity and artificial neural network to approximate the ...
Koroteev, D., Muravleva, E., Oseledets, I.   +2 more
core   +1 more source

Multivariate Neural Network Operators: Simultaneous Approximation and Voronovskaja‐Type Theorem

Mathematical Methods in the Applied Sciences
ABSTRACTIn this paper, the simultaneous approximation and a Voronoskaja‐type theorem for the multivariate neural network operators of the Kantorovich type have been proved. In order to establish such results, a suitable multivariate Strang–Fix type condition has been assumed.
Cantarini M., Costarelli D.
openaire   +3 more sources

Heuristically Adaptive Diffusion‐Model Evolutionary Strategy

Advanced Science, EarlyView.
Building on the mathematical equivalence between diffusion models and evolutionary algorithms, researchers demonstrate unprecedented control over evolutionary optimization through conditional diffusion. By training diffusion models to associate parameters with specific traits, they can guide evolution toward solutions exhibiting desired behaviors ...
Benedikt Hartl, Yanbo Zhang, Hananel Hazan, Michael Levin   +3 more
wiley   +1 more source

Coincidence Detection Using Spiking Neurons with Application to Face Recognition

Journal of Applied Mathematics, 2015
We elucidate the practical implementation of Spiking Neural Network (SNN) as local ensembles of classifiers. Synaptic time constant τs is used as learning parameter in representing the variations learned from a set of training data at classifier level ...
Fadhlan Kamaruzaman, Amir Akramin Shafie, Yasir M. Mustafah   +2 more
doaj   +1 more source

Hierarchical Summary Statistics Encoding Across Primary Visual and Posterior Parietal Cortices

Advanced Science, EarlyView.
This study shows that mouse V1 simultaneously encodes the ensemble mean and variance of motion, providing a robust summary‐statistic representation that persists despite single‐neuron variability. These signals propagate to PPC, where they are transformed into abstract category representations during decision making.
Young‐Beom Lee, Oliver James, Gaeun Jung, Doyun Lee, Yee‐Joon Kim   +4 more
wiley   +1 more source

The error-bounded descriptional complexity of approximation networks [PDF]

, 2010
It is well known that artificial neural nets can be used as approximators of any continuous functions to any desired degree and therefore be used e.g. in high - speed, real-time process control.
Brause, Rüdiger W.
core  

A wavelet neural network for the approximation of nonlinear multivariable function

IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028), 2000
Wavelet neural networks employing the wavelet function as the activation function have been proposed previously as an alternative approach to nonlinear mapping problems. In this paper, we propose a wavelet neural network which can be employed as a useful tool for learning a mapping between an input and an output space.
Ting Wang, Yasuo Sugai
openaire   +2 more sources

High‐Fidelity Synthetic Data Replicates Clinical Prediction Performance in a Million‐Patient Diabetes Cohort

Advanced Science, EarlyView.
This study generates high‐fidelity synthetic longitudinal records for a million‐patient diabetes cohort, successfully replicating clinical predictive performance. However, deeper analysis reveals algorithmic biases and trajectory inconsistencies that escape standard quality metrics. These findings challenge current validation norms, demonstrating why a
Francisco Ortuño, Víctor M. de la Oliva Roque, Javier‐Ignacio Ramirez‐Lopez, David P. Kreil, Joaquín Dopazo, Carlos Loucera   +5 more
wiley   +1 more source

Linearizing and Forecasting: A Reservoir Computing Route to Digital Twins of the Brain

Advanced Science, EarlyView.
A new approach uses simple neural networks to create digital twins of brain activity, capturing how different patterns unfold over time. The method generates and recovers key dynamics even from noisy data. When applied to fMRI, it predicts brain signals and reveals distinctive activity patterns across regions and individuals, opening possibilities for ...
Gabriele Di Antonio, Tommaso Gili, Andrea Gabrielli, Maurizio Mattia   +3 more
wiley   +1 more source