In-memory computing is an emerging non-von Neumann approach in which certain computational tasks such as matrix-vector multiplication are performed using resistive memory devices organized in a crossbar array. However, the conductance variations associated with the memory devices limit the precision of this computation. Here, we demonstrate that the so-called projected phase-change memory (Proj-PCM) devices can achieve 8-bit precision while performing scalar multiplication. The devices were fabricated and characterized using electrical measurements and STEM investigation. They are found to be remarkably immune to conductance variations arising from structural relaxation, 1/f noise and temperature variations. Moreover, it is possible to compensate for the temperature-dependent conductance variations in a crossbar array using a simple model. Finally, we experimentally demonstrate a neural network-based image classification task involving 30 such Proj-PCM devices.
DOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the "Taverne" license above, please follow below link for the End User Agreement:
Laboratory mice are predominantly used for one experiment only, i.e., new mice are ordered or bred for every new experiment. Moreover, most experiments use relatively young mice in the range of late adolescence to early adulthood. As a consequence, little is known about the day-to-day life of adult and aged laboratory mice. Here we present a long-term data set with three consecutive phases conducted with the same male mice over their lifetime in order to shed light on possible long-term effects of repeated cognitive stimulation. One third of the animals was trained by a variety of learning tasks conducted up to an age of 606 days. The mice were housed in four cages with 12 animals per cage; only four mice per cage had to repeatedly solve cognitive tasks for getting access to water using the IntelliCage system. In addition, these learner mice were tested in standard cognitive tests outside their home-cage. The other eight mice served as two control groups living in the same environment but without having to solve tasks for getting access to water. One control group was additionally placed on the test set-ups without having to learn the tasks. Next to the cognitive tasks, we took physiological measures (body mass, resting metabolic rate) and tested for dominance behavior, and attractivity in a female choice experiment. Overall, the mice were under surveillance until they died a natural death, providing a unique data set over the course of virtually their entire lives. Our data showed treatment differences during the first phase of our lifetime data set. Young learner mice showed a higher activity, less growth and resting metabolic rate, and were less attractive for female mice. These effects, however, were not preserved over the long-term. We also did not find differences in dominance or effects on longevity. However, we generated a unique and valuable set of long-term behavioral and physiological data from a single group of male mice and note that our long-term data contribute to a better understanding of the behavioral and physiological processes in male C57Bl/6J mice.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.