Results 101 to 110 of about 1,108,181 (276)
Memory‐Reduced Convolutional Neural Network for Fast Phase Hologram Generation
This article reports a lightweight convolutional neural network framework using INT8 quantization to efficiently generate 3D computer‐generated holograms from a single 2D image. The quantized model reduces memory usage and computational cost, accelerates inference speed, and maintains high output quality, enabling real‐time holographic display on low ...
Chenliang Chang +6 more
wiley +1 more source
This study introduces the first inverse machine learning model to predict laser powder bed fusion process parameters for targeted surface roughness of Inconel 718 parts. Unlike prior approaches, it incorporates spatial surface characteristics for improved accuracy.
Samsul Mahmood, Bart Raeymaekers
wiley +1 more source
Query DAGs: A Practical Paradigm for Implementing Belief-Network Inference
We describe a new paradigm for implementing inference in belief networks, which consists of two steps: (1) compiling a belief network into an arithmetic expression called a Query DAG (Q-DAG); and (2) answering queries using a simple evaluation algorithm.
Darwiche, A., Provan, G.
core +3 more sources
Enumeration of three term arithmetic progressions in fixed density sets [PDF]
Additive combinatorics is built around the famous theorem by Szemer\'edi which asserts existence of arithmetic progressions of any length among the integers. There exist several different proofs of the theorem based on very different techniques.
Sjöland, Erik
core
Aiming at the shortcomings of arithmetic optimization algorithm (AOA), which has low efficiency and is prone to fall into local optimal solutions, this paper proposes an improved AOA, called IAOA, and applies it to multi-threshold image segmentation ...
Hanyu Li +4 more
semanticscholar +1 more source
From Droplet to Diagnosis: Spatio‐Temporal Pattern Recognition in Drying Biofluids
This article integrates machine learning (ML) with the spatio‐temporal evolution of biofluid droplets to reveal how drying and self‐assembly encode distinctive compositional fingerprints. By leveraging textural features and interpretable ML, it achieves robust classification of blood abnormalities with over 95% accuracy.
Anusuya Pal +2 more
wiley +1 more source
Arithmetic coder optimization for compressing images obtained through remote probing of water bodies
Introduction. The fast program algorithm of arithmetic coding proposed in the paper is for the compression of digital images. It is shown how the complexity of the arithmetic coder algorithm depends on the complexity measures (the input size is not ...
R. V. Arzumanyan
doaj +1 more source
A multi-strategy enhanced arithmetic optimization algorithm called MSEAOA is proposed to address the issues of low population diversity, imbalanced exploration and exploitation capabilities, and low accuracy of optimal solution in the Arithmetic ...
Xuzhen Deng, Deng-xu He, Liangdong Qu
semanticscholar +1 more source
Quantization‐aware training creates resource‐efficient structured state space sequential S4(D) models for ultra‐long sequence processing in edge AI hardware. Including quantization during training leads to efficiency gains compared to pure post‐training quantization.
Sebastian Siegel +5 more
wiley +1 more source
Argon Cold Plasma Modification of Polyethylene Films to Tailor Surface and Barrier Properties
Argon cold plasma treatment of low‐density polyethylene (LDPE) and high‐density polyethylene (HDPE) films induces major changes in surface chemistry and morphology. Contact angles decrease significantly while surface energy increases, improving wettability. FTIR confirms the formation of polar functional groups, while pronounced etching, roughness, and
Aikaterini Spanou +11 more
wiley +1 more source

