Results 41 to 50 of about 1,244,429 (252)

Evaluation of Empirical Strategies to Treat Correlated and Heteroscedastic Noise in Multivariate Chemical Measurements [PDF]

open access: yes, 2019
The analysis of multivariate chemical data is often complicated by the presence of errors which are correlated or have non-uniform variance (heteroscedastic).
Giglio, Cannon
core   +1 more source

Label‐Free Assessment of Mesenchymal Stem Cell Quality by AI‐Driven Analysis of High‐Throughput Morphological Diversity

open access: yesAdvanced Science, EarlyView.
This label‐free AI‐powered workflow evaluates mesenchymal stem cell quality using high‐throughput flow‐cytometry light‐scatter morphological signals. It predicts cell potency and enables real‐time harvest inspection, cell‐batch evaluation, and in vivo therapeutic efficacy forecasting.
Di Wu   +10 more
wiley   +1 more source

Flexible Ultralow‐Power IGZO Multimodal Synaptic Transistors for Thermally Robust Neuromorphic Vision and Wearable Heatstroke Monitoring

open access: yesAdvanced Science, EarlyView.
A flexible, energy‐efficient amorphous IGZO‐based optoelectronic synaptic transistor is developed for multimodal neuromorphic sensing, enabling simultaneous processing of optical and thermal stimuli. The device exhibits ultra‐low power operation (1.6 fJ per synaptic event), stable performance under bending, and efficient optical signal processing.
Manoj Kumar Rajbhar   +4 more
wiley   +1 more source

Information Directed Sampling and Bandits with Heteroscedastic Noise

open access: yes, 2018
Proceedings of Machine Learning Research ...
Kirschner, Johannes, Krause, Andreas
openaire   +4 more sources

Tests of Unit Root Hypothesis With Heavy-Tailed Heteroscedastic Noises

open access: yesStatistica Sinica, 2023
Summary: This study examines unit-root testing with unspecified and heavy-tailed heteroscedastic noise. A new weighted least squares estimation (WLSE) is designed for the Dickey-Fuller (DF) test, the asymptotic normality of which is verified. However, the performance of the DF test relies strongly on the estimation accuracy of the asymptotic variance ...
openaire   +2 more sources

A human‐in‐the‐loop Bayesian optimization framework for constraint‐aware bioprocess development

open access: yesAIChE Journal, EarlyView.
Abstract We present Pareto Front Guided Sampling (PFGS), a human‐in‐the‐loop (HiL) Bayesian Optimization (BO) framework that reformulates surrogate‐model predictions as a multi‐objective optimization problem, allowing domain knowledge to be added intuitively to any BO problem.
Samuel Stricker   +6 more
wiley   +1 more source

Probabilistic wind power modelling via heteroscedastic non-stationary gaussian processes

open access: yesEnergy Reports
Accurate probabilistic prediction of wind power is crucial for maintaining grid stability and facilitating the efficient integration of renewable energy sources.
Domniki Ladopoulou   +2 more
doaj   +1 more source

Genetic Algorithm Approach to the 3D Node Localization in TDOA Systems

open access: yesSensors, 2019
Positioning asynchronous architectures based on time measurements are reaching growing importance in Local Positioning Systems (LPS). These architectures have special relevance in precision applications and indoor/outdoor navigation of automatic vehicles
Javier Díez-González   +5 more
doaj   +1 more source

Particle Swarm Optimization as a Noise‐Tolerant Strategy for High‐Entropy Alloy Research

open access: yesAdvanced Intelligent Discovery, EarlyView.
A customized particle swarm optimization algorithm is benchmarked against batch Bayesian optimization across four high‐entropy alloy oxygen‐reduction landscapes. While both methods locate the global optimum comparably under noise‐free conditions, Bayesian optimization’s surrogate model tends to overfit noisy data.
Ahmad Tirmidzi   +2 more
wiley   +1 more source

Innovation Pathways to Carbon Efficiency: Disentangling the Effects of AI, R&D, and Clean Energy Blessings on U.S. Environmental Sustainability

open access: yesBusiness Strategy and the Environment, EarlyView.
ABSTRACT The United States (U.S.) faces challenges in achieving its ambitious net‐zero carbon emissions target by 2050, with current emissions having fallen by less than 1% in 2024. Despite an investment of $500 billion in low‐carbon resources while holding the second‐largest green technology patent portfolio globally, it is further imperative to ...
Md Zubair Ahmad   +5 more
wiley   +1 more source

Home - About - Disclaimer - Privacy