Results 41 to 50 of about 4,893,936 (283)

Hybrid Precision Floating-Point (HPFP) Selection to Optimize Hardware-Constrained Accelerator for CNN Training

Sensors
The rapid advancement in AI requires efficient accelerators for training on edge devices, which often face challenges related to the high hardware costs of floating-point arithmetic operations.
Muhammad Junaid, Hayotjon Aliev, SangBo Park, HyungWon Kim, Hoyoung Yoo, Sanghoon Sim   +5 more
doaj   +1 more source

Hardware Acceleration of Stochastic Neural Networks Enabled by Bit-Cell Level Co-Design of Magnetic Tunnel Junctions. [PDF]

Adv Sci (Weinh)
An MTJ‐based probabilistic processing‐in‐memory architecture unites nonvolatile storage, local entropy generation, and stochastic computation within each memory cell. Stored multi‐bit weights are converted directly into programmable stochastic bitstreams, enabling neural‐network inference, adaptive Ising optimization, and uncertainty‐aware Bayesian ...
Wang Q, Koh D, McGoldrick B, Mocioi T, Fan Y, Baldo M, Liu L.   +6 more
europepmc   +2 more sources

Simulation‐Based Analysis of Insert Pull‐Out in Nickel‐Polyurethane Hybrid Foams Using CT‐Derived Geometries

Advanced Engineering Materials, EarlyView.
CT‐based finite element simulations combined with in situ X‐ray computed tomography are used to analyze insert pull‐out in nickel‐coated polymer foams. Despite variations in material parameters, deformation consistently concentrates within a narrow annular region around the insert.
Yannik Bautz, Jonas Fell, Hendrik Jost, Stefan Diebels, Hans‐Georg Herrmann   +4 more
wiley   +1 more source

Morphology, Transport, and Dynamics of Protein Adsorption in Open‐Cell Metal Foam

Advanced Engineering Materials, EarlyView.
Stainless steel (SS) open‐cell foams are shown to adsorb more protein per unit area than previously reported 316L SS and chromium oxide surfaces under static and flow conditions. An integrated approach combining 3D pore imaging, flow simulation, and protein adsorption experiments characterizes the foam’s performance.
Chinmaya Prerana Inguva, Joseph Lavoie, Afsaneh Rabiei   +2 more
wiley   +1 more source

Optimizing FDTD Memory Bandwidth by Using Block Float-Point Arithmetic

Elektronika ir Elektrotechnika, 2018
Finite-difference time-domain is a numerical method used for modelling of computational electrodynamics. The method is resource intensive, especially regarding memory usage. Multiple memory accesses are required per single computation so memory bandwidth
Stefan Pijetlovic, Milos Subotic, Nebojsa Pjevalica   +2 more
doaj   +1 more source

Improved architectures for fused floating-point arithmetic units [PDF]

, 2013
textMost general purpose processors (GPP) and application specific processors (ASP) use the floating-point arithmetic due to its wide and precise number system.
Sohn, Jongwook
core  

Framework for rapid hardware prototyping using custom floating-point arithmetic

, 2022
Framework for rapid hardware prototyping using custom floating-point arithmetic 
Nelson De-Sousa-Campos (5686436)
core   +1 more source

A Novel Approach to Estimate the Transition Temperature via Dynamic Nanoindentation

Advanced Engineering Materials, EarlyView.
A new dynamic nanoindentation‐based method was developed that uses the stiffness ratio as an indicator of the elastic–plastic deformation contributions at different temperatures. The approach successfully identified transition temperatures in ferritic steel and distinguished them from continuously ductile austenitic steel.
Stefan Zeiler, Nikolaus Kostwein, Ronald Schnitzer, Christian Mitterer, Verena Maier‐Kiener   +4 more
wiley   +1 more source

Improved Real-time Implementation of Adaptive Gassian Mixture Model-based Object Detection Algorithm for Fixed-point DSP Processors

Journal of Measurement Science and Instrumentation, 2010
Foreground moving object detection is an important process in various computer vision applications such as intelligent visual surveillance,HCI,object-based video compression,etc.One of the most successful moving object detection algorithms is based on ...
Byung-eun LEE, Thanh-binh NGUYEN, Sun-tae CHUNG   +2 more
doaj  

Hardware‐Native Gaussian Sampling With Probabilistic Ising Machines for Modern Communication Paradigms [PDF]

Advanced science
Probabilistic Ising machines are introduced as hardware‐native Gaussian samplers for edge communication workloads. Tunable variance and sampling density enable accurate distributions for Falcon‐class post‐quantum signatures and diffusion‐based semantic communication.
Rodrigues D, Grimaldi A, Volpe D, Bao Y, Yang S, Grassucci E, Siracusano G, Chiappini M, Carpentieri M, Yang H, Finocchio G.   +10 more
europepmc   +2 more sources