Zobrazeno 1 - 10
of 60
pro vyhledávání: '"Xiaochen Peng"'
Publikováno v:
Cost Effectiveness and Resource Allocation, Vol 21, Iss 1, Pp 1-5 (2023)
Abstract In recent years, international academics recognized that quality-adjusted life-years (QALYs) may not always fully capture the benefits produced by an intervention, and considered incorporating additional elements of value into cost-effective
Externí odkaz:
https://doaj.org/article/0825526c018f47cda33cef7a7a452c4e
Publikováno v:
Cost Effectiveness and Resource Allocation, Vol 21, Iss 1, Pp 1-5 (2023)
Abstract The use of multiple cost-effectiveness thresholds in pharmacoeconomic evaluation is a hotly debated topic in the international academic community. This study analyzed and discussed thresholds in the context of pharmacoeconomic evaluation and
Externí odkaz:
https://doaj.org/article/cef180059a964e3f8d4b416638f0318c
Publikováno v:
Cost Effectiveness and Resource Allocation, Vol 21, Iss 1, Pp 1-11 (2023)
Abstract The objective of this study was to estimate the willingness to pay (WTP) per quality-adjusted life year (QALY) among people with malignancies in China. The WTP for a QALY was estimated using a contingent valuation survey. Health utility was
Externí odkaz:
https://doaj.org/article/141369c5074343c7b88a6144a3e5197d
Publikováno v:
Frontiers in Artificial Intelligence, Vol 4 (2021)
Compute-in-memory (CIM) is an attractive solution to process the extensive workloads of multiply-and-accumulate (MAC) operations in deep neural network (DNN) hardware accelerators. A simulator with options of various mainstream and emerging memory te
Externí odkaz:
https://doaj.org/article/8ac8edafb035461bb11a2448dca72775
Publikováno v:
IEEE Journal on Exploratory Solid-State Computational Devices and Circuits, Vol 5, Iss 2, Pp 142-150 (2019)
In-memory computing with analog nonvolatile memories can accelerate the in situ training of deep neural networks. Recently, we proposed a synaptic cell of a ferroelectric transistor (FeFET) with two CMOS transistors (2T1F) that exploit the hybrid pre
Externí odkaz:
https://doaj.org/article/f842638bb9ae44349acdbda4ad5f4d45
Autor:
Ankit Kaul, Yandong Luo, Xiaochen Peng, Madison Manley, Yuan-Chun Luo, Shimeng Yu, Muhannad S. Bakir
Publikováno v:
IEEE Transactions on Electron Devices. 70:485-492
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0dea8dcd2410bc0a3c4e79545399769e
https://doi.org/10.1109/ted.2022.3231570
https://doi.org/10.1109/ted.2022.3231570
Publikováno v:
ACM Transactions on Design Automation of Electronic Systems. 27:1-19
On-device embedded artificial intelligence prefers the adaptive learning capability when deployed in the field, and thus in situ training is required. The compute-in-memory approach, which exploits the analog computation within the memory array, is a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::101cd7c914bd3dcadecd8ea35c10bbd5
https://doi.org/10.1145/3500929
https://doi.org/10.1145/3500929
Publikováno v:
IEEE Transactions on Electron Devices. 68:5598-5605
Emerging nonvolatile memory (eNVM)-based compute-in-memory (CIM) accelerators have been proven in silicon for machine learning at the macrolevel. To fully unleash the system-level benefits, the heterogeneous 3-D integration (H3D) using through-silico
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::103f14f7d59475dfbc3a3cd66d358ac0
https://doi.org/10.1109/ted.2021.3111857
https://doi.org/10.1109/ted.2021.3111857
Publikováno v:
IEEE Transactions on Circuits and Systems I: Regular Papers. 68:2753-2765
To efficiently deploy machine learning applications to the edge, compute-in-memory (CIM) based hardware accelerator is a promising solution with improved throughput and energy efficiency. Instant-on inference is further enabled by emerging non-volati
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::80bf079c49c05da4cac081b295469231
https://doi.org/10.1109/tcsi.2021.3072200
https://doi.org/10.1109/tcsi.2021.3072200
Publikováno v:
ACM Transactions on Design Automation of Electronic Systems. 26:1-18
Compute-in-memory (CIM) is an attractive solution to address the “memory wall” challenges for the extensive computation in deep learning hardware accelerators. For custom ASIC design, a specific chip instance is restricted to a specific network d
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3bc9e612d64e135b8d307e8f2e521452
https://doi.org/10.1145/3460436
https://doi.org/10.1145/3460436