Results 71 to 80 of about 10,729 (204)

Balancing Reconstruction Error and Kullback-Leibler Divergence in Variational Autoencoders

IEEE Access, 2020
Likelihood-based generative frameworks are receiving increasing attention in the deep learning community, mostly on account of their strong probabilistic foundation.
Andrea Asperti, Matteo Trentin
doaj   +1 more source

Collaborative Visual Localization for Modular Self‐Reconfigurable Robots

Advanced Intelligent Systems, EarlyView.
Relative localization in modular self‐reconfigurable robots is challenged by hardware limitations, constrained fields of view, and sensor faults. This paper, based on the SnailBot platform, presents a vision‐based collaborative localization method that combines ArUco markers with learning‐based algorithms to enable robust pose estimation from ...
Guanqi Liang, Rong Ou, Peiqi Wang, Xianghong Wang, Tin Lun Lam   +4 more
wiley   +1 more source

Notes on Kullback-Leibler Divergence and Likelihood

CoRR, 2014
The Kullback-Leibler (KL) divergence is a fundamental equation of information theory that quantifies the proximity of two probability distributions. Although difficult to understand by examining the equation, an intuition and understanding of the KL divergence arises from its intimate relationship with likelihood theory.
openaire   +2 more sources

Homophily‐adjusted social influence estimation

Canadian Journal of Statistics, EarlyView.
Abstract Homophily and social influence are two key concepts of social network analysis. Distinguishing between these phenomena is difficult, and approaches to disambiguate the two have been primarily limited to longitudinal data analyses. In this study, we provide sufficient conditions for valid estimation of social influence through cross‐sectional ...
Hanh T.D. Pham, Daniel K. Sewell
wiley   +1 more source

The McMillan Theorem for Colored Branching Processes and Dimensions of Random Fractals

Entropy, 2014
For the simplest colored branching process, we prove an analog to the McMillan theorem and calculate the Hausdorff dimensions of random fractals defined in terms of the limit behavior of empirical measures generated by finite genetic lines.
Victor Bakhtin
doaj   +1 more source

A generalization of the Kullback–Leibler divergence and its properties [PDF]

Journal of Mathematical Physics, 2009
A generalized Kullback–Leibler relative entropy is introduced starting with the symmetric Jackson derivative of the generalized overlap between two probability distributions. The generalization retains much of the structure possessed by the original formulation.
openaire   +3 more sources

Surface Consistent Seismic Hazard Assessment in the Shillong Plateau, Northeast India: A Probabilistic Approach Using Multiple Source Models and a Logic Tree Framework

Earthquake Engineering &Structural Dynamics, EarlyView.
ABSTRACT We perform probabilistic seismic hazard analysis (PSHA) in the Shillong Plateau by clubbing the up‐to‐date earthquake data and seismotectonic properties of the study area. Twenty‐one areal seismic sources are delineated based on the updated earthquake catalog, tectonic, and fault plane solutions.
Mohd Shahabuddin, William Kumar Mohanty, Eleftheria Papadimitriou   +2 more
wiley   +1 more source

Artificial intelligence–driven decoupling structure–activity relationship for lithium‐ion batteries

InfoScience, EarlyView.
Artificial intelligence can efferently accelerate the high‐throughput screening of battery materials, the analysis of multiphase mechanisms, and the precise prediction of capacity and cycle life. This review systematically summarizes the applications of machine learning (ML) in decoupling the complex structure‐activity relationships of lithium‐ion ...
Tao Wang, Yuhang Meng, Tianhao Chen, Xiangfen Jiang, Yoshio Bando, Jinjue Zeng, Xuebin Wang   +6 more
wiley   +1 more source

Physics‐Informed Generative Machine Learning for Designing Crack‐Free γ′‐Strengthened Ni‐Based Superalloys for Laser Powder Bed Fusion

Materials Genome Engineering Advances, EarlyView.
This research proposes a physics‐informed generative machine learning framework to design SHA800, a crack‐free γ′‐strengthened nickel‐based superalloy for laser powder bed fusion, achieving a 43% γ′ volume fraction and 587 HV0.2 hardness. ABSTRACT Fabricating γ′‐strengthened nickel‐based superalloys via laser powder bed fusion (LPBF) faces significant ...
Kai Guo, Dongwei Guan, Tao Hu, Yaqing Hou, Xiaoqun Li, Zhongnan Bi, Songzhe Xu, Hao Yu, Wei Xu, Weidong Xuan, Chaoyue Chen, Zhongming Ren   +11 more
wiley   +1 more source

Generative MR Multitasking With Complex‐Harmonic Cardiac Encoding: Bridging the Gap Between Gated Imaging and Real‐Time Imaging

Magnetic Resonance in Medicine, EarlyView.
ABSTRACT Purpose To develop a unified image reconstruction framework that bridges real‐time and gated cardiac MRI, including quantitative MRI. Methods We introduce generative multitasking, which learns subject‐ and dataset‐specific implicit neural temporal bases from sequence timings and an interpretable latent space for cardiac and respiratory motion.
Xinguo Fang, Anthony G. Christodoulou
wiley   +1 more source