Results 151 to 160 of about 33,813 (262)

Toward Capacitive In‐Memory‐Computing: A Device to Systems Level Perspective on the Future of Artificial Intelligence Hardware

open access: yesAdvanced Intelligent Discovery, EarlyView.
Capacitive, charge‐domain compute‐in‐memory (CIM) stores weights as capacitance,eliminating DC sneak paths and IR‐drop, yielding near‐zero standbypower. In this perspective, we present a device to systems level performance analysis of most promising architectures and predict apathway for upscaling capacitive CIM for sustainable edge computing ...
Kapil Bhardwaj   +2 more
wiley   +1 more source

A Flexible and Energy‐Efficient Compute‐in‐Memory Accelerator for Kolmogorov–Arnold Networks

open access: yesAdvanced Intelligent Systems, EarlyView.
This article presents KA‐CIM, a compute‐in‐memory accelerator for Kolmogorov–Arnold Networks (KANs). It enables flexible and efficient computation of arbitrary nonlinear functions through cross‐layer co‐optimization from algorithm to device. KA‐CIM surpasses CPU, ASIC, VMM‐CIM, and prior KAN accelerators by 1–3 orders of magnitude in energy‐delay ...
Chirag Sudarshan   +6 more
wiley   +1 more source

Worm‐Inspired Soft Robots With Modular Outfit‐Changing for Intelligent Multienvironment Adaptation

open access: yesAdvanced Intelligent Systems, EarlyView.
This study proposes a worm‐inspired soft robot capable of locomotion across multiple environments through a modular “outfit‐changing” strategy. The robot integrates pneumatically actuated peristaltic segments with interchangeable external modules, enabling efficient motion on ground surfaces, within pipelines, through granular media, and underwater ...
Xiaomin Liu   +6 more
wiley   +1 more source

Voronoi Tessellations and the Shannon Entropy of the Pentagonal Tilings. [PDF]

open access: yesEntropy (Basel), 2023
Bormashenko E   +4 more
europepmc   +1 more source

Methods for Setting Device Specifications for Analog In‐Memory Computing Inference

open access: yesAdvanced Intelligent Systems, EarlyView.
Non‐volatile memories (NVMs) are being developed for analog in‐memory computing for energy‐efficient, high‐speed deep learning inference. As technology is moving to industry adoption, a method to define required NVM specifications is critical for improving performance and reducing manufacturing cost.
Zhenyu Wu   +3 more
wiley   +1 more source

Tiled material systems: Exploring biodiversity and multifunctionality of a universal and structural motif. [PDF]

open access: yesPNAS Nexus
Ciecierska-Holmes J   +8 more
europepmc   +1 more source

Home - About - Disclaimer - Privacy