Results 71 to 80 of about 41,144 (258)

Subspace Inference for Bayesian Deep Learning

CoRR, 2019
Bayesian inference was once a gold standard for learning with neural networks, providing accurate full predictive distributions and well calibrated uncertainty. However, scaling Bayesian inference techniques to deep neural networks is challenging due to the high dimensionality of the parameter space.
Pavel Izmailov, Wesley J. Maddox, Polina Kirichenko, Timur Garipov, Dmitry P. Vetrov, Andrew Gordon Wilson   +5 more
openaire   +4 more sources

Electrolyte Engineering Challenges and Opportunities for Next‐Generation Aqueous Ammonium‐Ion Batteries

Advanced Materials, EarlyView.
Aqueous ammonium‐ion batteries (AAIBs) face a hydrogen‐bond paradox: the HB network enables fast NH4+ transport but triggers water decomposition. This review dissects this dilemma, evaluates multiple electrolyte engineering strategies, and outlines four future directions for next‐generation AAIB design ABSTRACT Aqueous ammonium‐ion batteries (AAIBs ...
Zi‐Hang Huang, Xiaoyu Liu, Shuai Mao, Lingfeng Zhu, Zhijun Wu, Hongge Pan, Tianyi Ma   +6 more
wiley   +1 more source

Deep learning solutions for smart city challenges in urban development

Scientific Reports
In the realm of urban planning, the integration of deep learning technologies has emerged as a transformative force, promising to revolutionize the way cities are designed, managed, and optimized.
Pengjun Wu, Zhanzhi Zhang, Xueyi Peng, Ran Wang   +3 more
doaj   +1 more source

Blast Loading Prediction of Complex Structures Based on Bayesian Deep Active Learning

Applied Sciences
The prediction of blast loading for complex structures using deep learning requires extensive training data from field experiments or numerical simulations.
Meilin Pan, Weiwen Peng, Chunjiang Leng, Jiulu Qiu, Wei Zhong   +4 more
doaj   +1 more source

Bioengineered Interfaces for Peripheral Nerve Sensory Restoration

Advanced Materials, EarlyView.
Half of amputees abandon their prosthetics for lack of feeling. This review charts the full path from peripheral nerve injury to restored sensation, through surgical, regenerative, noninvasive, and implanted approaches, and shows how injury type and interface material properties determine which strategy can deliver naturalistic feedback, and why ...
Sydney Swedick, Salim El Hadwe, Kesi Liang, George G. Malliaras, Damiano G. Barone   +4 more
wiley   +1 more source

Maneuver strategy recognition technology for enemy combat aircraft based on Bayesian deep learning

Shenzhen Daxue xuebao. Ligong ban
Enhancing identification of enemy combat aircraft maneuver strategies is a critical factor in improving air combat decision-making capabilities. As traditional deep learning models often show overconfidence in complex and variable combat environment, and
YUAN Yinlong, ZHANG Sijie, CHENG Yun, HUA Liang   +3 more
doaj   +1 more source

Improving stroke diagnosis accuracy using hyperparameter optimized deep learning

IJAIN (International Journal of Advances in Intelligent Informatics), 2019
Stroke may cause death for anyone, including youngsters. One of the early stroke detection techniques is a Computerized Tomography (CT) scan. This research aimed to optimize hyperparameter in Deep Learning, Random Search and Bayesian Optimization for ...
Tessy Badriyah, Dimas Bagus Santoso, Iwan Syarif, Daisy Rahmania Syarif   +3 more
doaj   +1 more source

Generalized Bayesian deep reinforcement learning

CoRR
Bayesian reinforcement learning (BRL) is a method that merges principles from Bayesian statistics and reinforcement learning to make optimal decisions in uncertain environments. As a model-based RL method, it has two key components: (1) inferring the posterior distribution of the model for the data-generating process (DGP) and (2) policy learning using
Shreya Sinha Roy, Richard G. Everitt, Christian P. Robert, Ritabrata Dutta   +3 more
openaire   +2 more sources

Advanced MXene‐Based Multifunctional Nanoarchitecture Materials Engineered for Adsorptive Cleanup of Hazardous Radioactive Pollutants: A Comprehensive Critical Review

Advanced Materials Interfaces, EarlyView.
This work critically reviews MXenes as highly effective multifunctional nanomaterials for the adsorption of radio‐contaminants, demonstrating a remarkable adsorption capacity of up to 1376.75 mg/g and cyclic stability of 2–8 cycles, with complexation, electrostatic interactions, and the numerical strength of MXene active sites playing a key operational
Stephen Sunday Emmanuel, Shepherd M. Tichapondwa   +1 more
wiley   +1 more source

Hyperparameter optimization of machine learning models for predicting actual evapotranspiration

Machine Learning with Applications
Direct measurement of actual evapotranspiration (AET) using eddy covariance and lysimeters is challenging, particularly in large areas, due to high cost, technical complexity, and the need for specialized instrumentation.
Chalachew Muluken Liyew, Elvira Di Nardo, Stefano Ferraris, Rosa Meo   +3 more
doaj   +1 more source