Hyperdimensional Computing with Learnable Projection for User Adaptation Framework

التفاصيل البيبلوغرافية
العنوان: Hyperdimensional Computing with Learnable Projection for User Adaptation Framework
المؤلفون: Hsiao, Yu-Ren, Chuang, Yu-Chuan, Chang, Cheng-Yang, Wu, An-Yeu
المساهمون: National Taiwan University Taiwan (NTU), Ilias Maglogiannis, John Macintyre, Lazaros Iliadis, TC 12, WG 12.5
المصدر: IFIP Advances in Information and Communication Technology ; 17th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI) ; https://hal.inria.fr/hal-03287688Test ; 17th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), Jun 2021, Hersonissos, Crete, Greece. pp.436-447, ⟨10.1007/978-3-030-79150-6_35⟩
بيانات النشر: HAL CCSD
Springer International Publishing
سنة النشر: 2021
المجموعة: Archive ouverte HAL (Hyper Article en Ligne, CCSD - Centre pour la Communication Scientifique Directe)
مصطلحات موضوعية: Brain-inspired computing, Hyperdimensional computing, User adaptation, [INFO]Computer Science [cs]
جغرافية الموضوع: Hersonissos, Crete, Greece
الوصف: Part 11: Hyperdimensional Computing ; International audience ; Brain-inspired Hyperdimensional Computing (HDC), a machine learning (ML) model featuring high energy efficiency and fast adaptability, provides a promising solution to many real-world tasks on resource-limited devices. This paper introduces an HDC-based user adaptation framework, which requires efficient fine-tuning of HDC models to boost accuracy. Specifically, we propose two techniques for HDC, including the learnable projection and the fusion mechanism for the Associative Memory (AM). Compared with the user adaptation framework based on the original HDC, our proposed framework shows 4.8% and 3.5% of accuracy improvements on two benchmark datasets, including the ISOLET dataset and the UCIHAR dataset, respectively.
نوع الوثيقة: conference object
اللغة: English
العلاقة: hal-03287688; https://hal.inria.fr/hal-03287688Test; https://hal.inria.fr/hal-03287688/documentTest; https://hal.inria.fr/hal-03287688/file/509922_1_En_35_Chapter.pdfTest
DOI: 10.1007/978-3-030-79150-6_35
الإتاحة: https://doi.org/10.1007/978-3-030-79150-6_35Test
https://hal.inria.fr/hal-03287688Test
https://hal.inria.fr/hal-03287688/documentTest
https://hal.inria.fr/hal-03287688/file/509922_1_En_35_Chapter.pdfTest
حقوق: http://creativecommons.org/licenses/byTest/
رقم الانضمام: edsbas.3C89CA41
قاعدة البيانات: BASE