Results 71 to 80 of about 127,617 (265)
, 2017 We propose proximal backpropagation (ProxProp) as a novel algorithm that takes implicit instead of explicit gradient steps to update the network parameters during neural network training. Our algorithm is motivated by the step size limitation of explicit gradient descent, which poses an impediment for optimization.
Frerix, Thomas +3 more
openaire +2 more sources
Unveil Fundamental Graph Properties for Neural Architecture Search
Advanced Science, EarlyView.This paper proposes NASGraph, a graph‐based framework that represents neural architectures as graphs whose structural properties determine performance. By revealing structure–performance relationships, NASGraph enables efficient neural architecture search with significantly reduced computation.
Zhenhan Huang +4 more
wiley +1 more source
Revista Tecnológica, 2014 La aplicación de modelos matemáticos en el manejo de cuencas hidrográficas tiene requerimientos exigentes de información y en su mayoría no han sido desarrollados para ser aplicados en regiones de montaña.
Jaime Eduardo Veintimilla +1 more
doaj
The Prediction of Yarn Elongation of Kenyan Ring-Spun Yarn using Extreme Learning Machines (ELM) [PDF]
Tekstilec, 2017 The optimization of the manufacture of cotton yarns involves several processes, while the prediction of yarn quality parameters forms an important area of investigation. This research work concentrated on the prediction of cotton yarn elongation.
Josphat Igadwa Mwasiagi
doaj +1 more source
Improving neural networks by preventing co-adaptation of feature detectors [PDF]
, 2012 When a large feedforward neural network is trained on a small training set, it typically performs poorly on held-out test data. This "overfitting" is greatly reduced by randomly omitting half of the feature detectors on each training case.
Hinton, Geoffrey E. +4 more
core +1 more source
NanoLoop: A Deep Learning Framework Leveraging Nanopore Sequencing for Chromatin Loop Prediction
Advanced Science, EarlyView.Chromatin loops are central to gene regulation and 3D genome organization. Leveraging Nanopore sequencing's ability to jointly capture DNA sequence and methylation, we present NanoLoop, the first framework for genome‐wide chromatin loop prediction using Nanopore data.
Wenjie Huang +5 more
wiley +1 more source
A Novel Neuro-Fuzzy Model for Multivariate Time-Series Prediction
Data, 2018 Time series forecasting can be a complicated problem when the underlying process shows high degree of complex nonlinear behavior. In some domains, such as financial data, processing related time-series jointly can have significant benefits.
Alexander Vlasenko +3 more
doaj +1 more source
Generating Dynamic Structures Through Physics‐Based Sampling of Predicted Inter‐Residue Geometries
Advanced Science, EarlyView.While static structure prediction has been revolutionized, modeling protein dynamics remains elusive. trRosettaX2‐Dynamics is presented to address this challenge. This framework leverages a Transformer‐based network to predict inter‐residue geometric constraints, guiding conformation generation via physics‐based iterative sampling. The resulting method
Chenxiao Xiang +3 more
wiley +1 more source
Regular Cycles of Forward and Backward Signal Propagation in Prefrontal Cortex and in Consciousness
Frontiers in Systems Neuroscience, 2016 This paper addresses two fundamental questions:(1) Is it possible to develop mathematical neural network models which can explain and replicate the way in which higher-order capabilities like intelligence, consciousness, optimization and prediction ...
Paul John Werbos +1 more
doaj +1 more source
Analog hardware for delta-backpropagation neural networks [PDF]
, 1992 This is a fully parallel analog backpropagation learning processor which comprises a plurality of programmable resistive memory elements serving as synapse connections whose values can be weighted during learning with buffer amplifiers, summing circuits,
Eberhardt, Silvio P.
core +1 more source