Results 81 to 90 of about 1,059,681 (321)

An algebraic characterization of the optimum of regularized kernel methods [PDF]

open access: yesMachine Learning, 2009
The representer theorem for kernel methods states that the solution of the associated variational problem can be expressed as the linear combination of a finite number of kernel functions. However, for non-smooth loss functions, the analytic characterization of the coefficients poses nontrivial problems.
DINUZZO, FRANCESCO, DE NICOLAO, GIUSEPPE
openaire   +3 more sources

The Role of Theoretical Calculations for INVEST Systems: Complementarity Between Theory and Experiments and Rationalization of the Results

open access: yesAdvanced Optical Materials, EarlyView.
Theoretical lessons are key for molecules presenting an inverted singlet‐triplet excited state (e.g. S1 and T1) energy difference. This perspective provides a snapshot of the role played by calculations in last years, not only to anticipate experimental findings but also for driving high‐throughput virtual screenings, as well as the main challenge to ...
Ángel José Pérez‐Jiménez   +2 more
wiley   +1 more source

Spintronic Memtransistor Leaky Integrate and Fire Neuron for Spiking Neural Networks

open access: yesAdvanced Electronic Materials, EarlyView.
Spintronic memtransistor neurons based on domain walls enable energy‐efficient, field‐gated, and current‐controlled LIF functionality for neuromorphic computing, as demonstrated. When integrated into spiking neural network architectures, these devices achieve >96% pattern recognition accuracy, demonstrating high performance, scalability, and mem ...
Aijaz H. Lone   +7 more
wiley   +1 more source

Noncommutative spectral synthesis for the involutive Banach algebra associated with a topological dynamical system

open access: yes, 2012
If X is a compact Hausdorff space, supplied with a homeomorphism, then a crossed product involutive Banach algebra is naturally associated with these data. If X consists of one point, then this algebra is the group algebra of the integers. In this paper,
de Jeu, Marcel, Tomiyama, Jun
core   +1 more source

Recent Advancements in Topic Modeling Techniques for Healthcare, Bioinformatics, and Other Potential Applications

open access: yesAdvanced Intelligent Systems, EarlyView.
This article offers a comprehensive review of topic modeling techniques, tracing their evolution from inception to recent developments. It explores methods such as latent Dirichlet allocation, latent semantic analysis, non‐negative matrix factorization, probabilistic latent semantic analysis, Top2Vec, and BERTopic, highlighting their strengths ...
Pratima Kumari   +6 more
wiley   +1 more source

Strong endomorphism kernel property for monounary algebras [PDF]

open access: yesMathematica Bohemica, 2018
All monounary algebras which have strong endomorphism kernel property are described.
Emília Halušková
doaj   +1 more source

Relational Supersymmetry via the Dressing Field Method and Matter‐Interaction Supergeometric Framework

open access: yesAnnalen der Physik, EarlyView.
Applications of the Dressing Field Method are reviewed and further expanded to the very foundations of the supersymmetric framework, where it allows to build relational supersymmetric field theory. Furthermore, a novel approach is proposed giving a unified description of fermionic matter fields and bosonic gauge fields: a Matter‐Interaction ...
Jordan François, L. Ravera
wiley   +1 more source

On the solution of the convolution equation with a sum-difference kernel

open access: yesVestnik Samarskogo Gosudarstvennogo Tehničeskogo Universiteta. Seriâ: Fiziko-Matematičeskie Nauki, 2015
The paper deals with the integral equations of the second kind with a sumdifference kernel. These equations describe a series of physical processes in a medium with a reflective boundary.
Ani G Barseghyan
doaj   +1 more source

Transfer Learning‐Based Data‐Fusion Model Framework for State of Health Estimation of Power Battery Packs

open access: yesBattery Energy, EarlyView.
To handle nonlinearity and feature coupling, a data‐driven aging model is proposed, employing dual Gaussian Process Regressions and transfer learning to enhance model efficiency and accuracy. Adaptive filtering refines the model by integrating aging features and output capacity, resulting in a closed‐loop data fusion framework for SOH estimation ...
Zhiqiang Lyu   +3 more
wiley   +1 more source

Home - About - Disclaimer - Privacy