Results 31 to 40 of about 1,606,312 (290)
Hydrogen, 2023 A proton exchange membrane (PEM) electrolyzer is fed with water and powered by electric power to electrochemically produce hydrogen at low operating temperatures and emits oxygen as a by-product.
Mohammad Biswas +2 more
doaj +1 more source
Robustness and Generalization [PDF]
, 2010 We derive generalization bounds for learning algorithms based on their robustness: the property that if a testing sample is "similar" to a training sample, then the testing error is close to the training error.
A. Ben-Tal +57 more
core +2 more sources
Enabling Training of Neural Networks on Noisy Hardware
Frontiers in Artificial Intelligence, 2021 Deep neural networks (DNNs) are typically trained using the conventional stochastic gradient descent (SGD) algorithm. However, SGD performs poorly when applied to train networks on non-ideal analog hardware composed of resistive device arrays with non ...
Tayfun Gokmen
doaj +1 more source
Наука Красноярья, 2023 The article proposes the integration of neural networks as a parallel element base in telecommunications systems. In this case, their main advantage, the ability to self-study or adapt to external conditions, will be used.
Pavel A. Sakhnyuk, Tatyana I. Sakhnyuk
doaj +1 more source
How to shift bias: Lessons from the Baldwin effect [PDF]
, 1996 An inductive learning algorithm takes a set of data as input and generates a hypothesis as output. A set of data is typically consistent with an infinite number of hypotheses; therefore, there must be factors other than the data that determine the output
Turney, Peter D.
core +2 more sources
Spectral Algorithms for Supervised Learning [PDF]
Neural Computation, 2008 We discuss how a large class of regularization methods, collectively known as spectral regularization and originally designed for solving ill-posed inverse problems, gives rise to regularized learning algorithms. All of these algorithms are consistent kernel methods that can be easily implemented. The intuition behind their derivation is that the same
LO GERFO L. +4 more
openaire +3 more sources
IEEE Access, 2021 Both non-equal trail lengths and non-zero initial errors are practical challenges to learning control of robotic and mechatronic systems. Iterative learning to update input is still desired, because of the repetitive motion nature of the controlled ...
Mingxuan Sun +3 more
doaj +1 more source
Optimizing 0/1 Loss for Perceptrons by Random Coordinate Descent [PDF]
, 2007 The 0/1 loss is an important cost function for perceptrons. Nevertheless it cannot be easily minimized by most existing perceptron learning algorithms. In this paper, we propose a family of random coordinate descent algorithms to directly minimize the 0 ...
Li, Ling, Lin, Hsuan-Tien
core +1 more source
Instance-Based Learning Algorithms [PDF]
Machine Learning, 1991 Storing and using specific instances improves the performance of several supervised learning algorithms. These include algorithms that learn decision trees, classification rules, and distributed networks. However, no investigation has analyzed algorithms that use only specific instances to solve incremental learning tasks.
David W. Aha +2 more
openaire +3 more sources
Integrating LiDAR, Multispectral and SAR Data to Estimate and Map Canopy Height in Tropical Forests
Remote Sensing, 2019 Developing accurate methods to map vegetation structure in tropical forests is essential to protect their biodiversity and improve their carbon stock estimation. We integrated LIDAR (Light Detection and Ranging), multispectral and SAR (Synthetic Aperture
J. Camilo Fagua +4 more
doaj +1 more source