Results 41 to 50 of about 109,285 (289)

All‐in‐One Analog AI Hardware: On‐Chip Training and Inference with Conductive‐Metal‐Oxide/HfOx ReRAM Devices

Advanced Functional Materials, EarlyView.
An all‐in‐one analog AI accelerator is presented, enabling on‐chip training, weight retention, and long‐term inference acceleration. It leverages a BEOL‐integrated CMO/HfOx ReRAM array with low‐voltage operation (<1.5 V), multi‐bit capability over 32 states, low programming noise (10 nS), and near‐ideal weight transfer.
Donato Francesco Falcone, Victoria Clerico, Wooseok Choi, Tommaso Stecconi, Folkert Horst, Laura Bégon‐Lours, Matteo Galetta, Antonio La Porta, Nikhil Garg, Fabien Alibart, Bert Jan Offrein, Valeria Bragaglia   +11 more
wiley   +1 more source

A Pipeline-Based ODE Solving Framework

IEEE Access
The traditional parallel solving methods of ordinary differential equations (ODE) are mainly classified into task-parallelism, data-parallelism, and instruction-level parallelism.
Ruixia Cao, Shangjun Hou, Lin Ma
doaj   +1 more source

Quantum Physics and Computers [PDF]

, 1996
Recent theoretical results confirm that quantum theory provides the possibility of new ways of performing efficient calculations. The most striking example is the factoring problem.
Adriano Barenco, Barenco A., Biafore M., Cohen-Tannoudji C., Ekert A., Hardy G. H., Holevo A. S., Knuth D. E., Margolus N., Miller G. L., Palma G. M., Peres A., Rivest R., Smolin J. A., Steane A., Vedral V., Welsh D., Wilf H. S.   +17 more
core   +3 more sources

An Ultra‐Robust Memristor Based on Vertically Aligned Nanocomposite with Highly Defective Vertical Channels for Neuromorphic Computing

Advanced Functional Materials, EarlyView.
An ultra‐robust memristor based on SrTiO3‐CeO2 (S‐C) vertically aligned nanocomposite (VAN) achieving exceptional endurance of 1012 switching cycles via interface engineering. Artificial neural networks (ANNs) integrated with S‐C VAN memristors exhibit high training accuracy across multiple datasets.
Zedong Hu, Hongyi Dou, Jeremy B. Gan, Cun Li, Xuanyu Sheng, Juanjuan Lu, Abhijeet Choudhury, Jialong Huang, Yizhi Zhang, Zhongxia Shang, Ye Cao, Xinghang Zhang, Haiyan Wang   +12 more
wiley   +1 more source

How Many Cooks Spoil the Soup? [PDF]

, 2016
In this work, we study the following basic question: "How much parallelism does a distributed task permit?" Our definition of parallelism (or symmetry) here is not in terms of speed, but in terms of identical roles that processes have at the same time in
D Alistarh, D Angluin, D Angluin, D Doty, D Malkhi, D Peleg, H Attiya, H-L Chen, H-L Chen, I Chatzigiannakis, J Aspnes, J Czyzowicz, J Suomela, K-T Förster, M Yamashita, NA Lynch, O Michail, O Michail, P Flocchini, Y Afek   +19 more
core   +2 more sources

Ultrahigh‐Yield, Multifunctional, and High‐Performance Organic Memory for Seamless In‐Sensor Computing Operation

Advanced Functional Materials, EarlyView.
Molecular engineering of a nonconjugated radical polymer enables a significant enhancement of the glass transition temperature. The amorphous nature and tunability of the polymer, arising from its nonconjugated backbone, facilitates the fabrication of organic memristive devices with an exceptionally high yield (>95%), as well as substantial ...
Daeun Kim, Jaehyoung Ko, Namju Kim, Quynh H. Nguyen, Hoyeon Lee, Joohwan Eo, Seongeon Jin, Ji Eon Kwon, Seung‐Yeol Jeon, Youngdo Jeong, Dongseok Kwon, Sung Gap Im, Sanghan Lee, Byung Chul Jang, Yongho Joo   +14 more
wiley   +1 more source

Parallelism Strategies for Big Data Delayed Transfer Entropy Evaluation

Algorithms, 2019
Generated and collected data have been rising with the popularization of technologies such as Internet of Things, social media, and smartphone, leading big data term creation. One class of big data hidden information is causality.
Jonas R. Dourado, Jordão Natal de Oliveira Júnior, Carlos D. Maciel   +2 more
doaj   +1 more source

Memory and Parallelism Analysis Using a Platform-Independent Approach

, 2019
Emerging computing architectures such as near-memory computing (NMC) promise improved performance for applications by reducing the data movement between CPU and memory. However, detecting such applications is not a trivial task.
Awan, Ahsan Javed, Corda, Stefano, Corporaal, Henk, Jordans, Roel, Singh, Gagandeep, Stuijk, Sander   +5 more
core   +1 more source

Task decomposition using pattern distributor [PDF]

, 2004
In this paper, we propose a new task decomposition method for multilayered feedforward neural networks, namely Task Decomposition with Pattern Distributor in order to shorten the training time and improve the generalization accuracy of a network under ...
Bao, C, Guan, SU, Neo, T
core   +1 more source

Spectrally Tunable 2D Material‐Based Infrared Photodetectors for Intelligent Optoelectronics

Advanced Functional Materials, EarlyView.
Intelligent optoelectronics through spectral engineering of 2D material‐based infrared photodetectors. Abstract The evolution of intelligent optoelectronic systems is driven by artificial intelligence (AI). However, their practical realization hinges on the ability to dynamically capture and process optical signals across a broad infrared (IR) spectrum.
Junheon Ha, Yingshan Ma, Yong Nam An, Sung‐Un An, Hyeon Hak Jung, Suvi‐Tuuli Varjamo, Jiwon Yoo, Junho Min, Hanvit Kim, Faisal Ahmed, Sang Hoon Chae, Young Min Song, Weiwei Cai, Tawfique Hasan, Zhipei Sun, Dong‐Ho Kang, Hyeon‐Jin Shin, Yunyun Dai, Hoon Hahn Yoon   +18 more
wiley   +1 more source