001038062 001__ 1038062
001038062 005__ 20250203103308.0
001038062 0247_ $$2doi$$a10.48550/ARXIV.2410.09755
001038062 0247_ $$2doi$$a10.48550/arXiv.2410.09755
001038062 0247_ $$2datacite_doi$$a10.34734/FZJ-2025-01111
001038062 037__ $$aFZJ-2025-01111
001038062 1001_ $$0P:(DE-Juel1)192242$$aManea, Paul-Philipp$$b0$$eCorresponding author$$ufzj
001038062 245__ $$aGain Cell-Based Analog Content Addressable Memory for Dynamic Associative tasks in AI
001038062 260__ $$barXiv$$c2024
001038062 3367_ $$0PUB:(DE-HGF)25$$2PUB:(DE-HGF)$$aPreprint$$bpreprint$$mpreprint$$s1738246021_31383
001038062 3367_ $$2ORCID$$aWORKING_PAPER
001038062 3367_ $$028$$2EndNote$$aElectronic Article
001038062 3367_ $$2DRIVER$$apreprint
001038062 3367_ $$2BibTeX$$aARTICLE
001038062 3367_ $$2DataCite$$aOutput Types/Working Paper
001038062 520__ $$aAnalog Content Addressable Memories (aCAMs) have proven useful for associative in-memory computing applications like Decision Trees, Finite State Machines, and Hyper-dimensional Computing. While non-volatile implementations using FeFETs and ReRAM devices offer speed, power, and area advantages, they suffer from slow write speeds and limited write cycles, making them less suitable for computations involving fully dynamic data patterns. To address these limitations, in this work, we propose a capacitor gain cell-based aCAM designed for dynamic processing, where frequent memory updates are required. Our system compares analog input voltages to boundaries stored in capacitors, enabling efficient dynamic tasks. We demonstrate the application of aCAM within transformer attention mechanisms by replacing the softmax-scaled dot-product similarity with aCAM similarity, achieving competitive results. Circuit simulations on a TSMC 28 nm node show promising performance in terms of energy efficiency, precision, and latency, making it well-suited for fast, dynamic AI applications.
001038062 536__ $$0G:(DE-HGF)POF4-5234$$a5234 - Emerging NC Architectures (POF4-523)$$cPOF4-523$$fPOF IV$$x0
001038062 536__ $$0G:(BMBF)16ME0400$$aBMBF 16ME0400 - Verbundprojekt: Neuro-inspirierte Technologien der künstlichen Intelligenz für die Elektronik der Zukunft - NEUROTEC II - (16ME0400)$$c16ME0400$$x1
001038062 588__ $$aDataset connected to DataCite
001038062 650_7 $$2Other$$aEmerging Technologies (cs.ET)
001038062 650_7 $$2Other$$aFOS: Computer and information sciences
001038062 7001_ $$0P:(DE-Juel1)194421$$aLeroux, Nathan$$b1$$ufzj
001038062 7001_ $$0P:(DE-Juel1)188273$$aNeftci, Emre$$b2$$ufzj
001038062 7001_ $$0P:(DE-Juel1)188145$$aStrachan, John Paul$$b3$$ufzj
001038062 773__ $$a10.48550/arXiv.2410.09755
001038062 8564_ $$uhttps://doi.org/10.48550/arXiv.2410.09755
001038062 8564_ $$uhttps://juser.fz-juelich.de/record/1038062/files/2410.09755v1.pdf$$yOpenAccess
001038062 909CO $$ooai:juser.fz-juelich.de:1038062$$pdnbdelivery$$pdriver$$pVDB$$popen_access$$popenaire
001038062 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)192242$$aForschungszentrum Jülich$$b0$$kFZJ
001038062 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)194421$$aForschungszentrum Jülich$$b1$$kFZJ
001038062 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188273$$aForschungszentrum Jülich$$b2$$kFZJ
001038062 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188145$$aForschungszentrum Jülich$$b3$$kFZJ
001038062 9131_ $$0G:(DE-HGF)POF4-523$$1G:(DE-HGF)POF4-520$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5234$$aDE-HGF$$bKey Technologies$$lNatural, Artificial and Cognitive Information Processing$$vNeuromorphic Computing and Network Dynamics$$x0
001038062 9141_ $$y2024
001038062 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001038062 920__ $$lyes
001038062 9201_ $$0I:(DE-Juel1)PGI-14-20210412$$kPGI-14$$lNeuromorphic Compute Nodes$$x0
001038062 9201_ $$0I:(DE-Juel1)PGI-15-20210701$$kPGI-15$$lNeuromorphic Software Eco System$$x1
001038062 980__ $$apreprint
001038062 980__ $$aVDB
001038062 980__ $$aUNRESTRICTED
001038062 980__ $$aI:(DE-Juel1)PGI-14-20210412
001038062 980__ $$aI:(DE-Juel1)PGI-15-20210701
001038062 9801_ $$aFullTexts