Zobrazeno 1 - 10
of 154
pro vyhledávání: '"Hu, X. Sharon"'
Content addressable memory (CAM) stands out as an efficient hardware solution for memory-intensive search operations by supporting parallel computation in memory. However, developing a CAM-based accelerator architecture that achieves acceptable accur
Externí odkaz:
http://arxiv.org/abs/2403.03442
Autor:
Meng, Guangyu, Zhou, Ruyu, Liu, Liu, Liang, Peixian, Liu, Fang, Chen, Danny, Niemier, Michael, Hu, X. Sharon
Earth Mover's Distance (EMD) is an important similarity measure between two distributions, used in computer vision and many other application domains. However, its exact calculation is computationally and memory intensive, which hinders its scalabili
Externí odkaz:
http://arxiv.org/abs/2401.07378
Autor:
Shou, Shengxi, Liu, Che-Kai, Yun, Sanggeon, Wan, Zishen, Ni, Kai, Imani, Mohsen, Hu, X. Sharon, Yang, Jianyi, Zhuo, Cheng, Yin, Xunzhao
In this work, we propose SEE-MCAM, scalable and compact multi-bit CAM (MCAM) designs that utilize the three-terminal ferroelectric FET (FeFET) as the proxy. By exploiting the multi-level-cell characteristics of FeFETs, our proposed SEE-MCAM designs e
Externí odkaz:
http://arxiv.org/abs/2310.04940
Autor:
Xu, Yixin, Zhao, Zijian, Xiao, Yi, Yu, Tongguang, Mulaosmanovic, Halid, Kleimaier, Dominik, Duenkel, Stefan, Beyer, Sven, Gong, Xiao, Joshi, Rajiv, Hu, X. Sharon, Wen, Shixian, Rios, Amanda Sofie, Lekkala, Kiran, Itti, Laurent, Homan, Eric, George, Sumitha, Narayanan, Vijaykrishnan, Ni, Kai
Field Programmable Gate Array (FPGA) is widely used in acceleration of deep learning applications because of its reconfigurability, flexibility, and fast time-to-market. However, conventional FPGA suffers from the tradeoff between chip area and recon
Externí odkaz:
http://arxiv.org/abs/2212.00089
Automatic classification of pigmented, non-pigmented, and depigmented non-melanocytic skin lesions have garnered lots of attention in recent years. However, imaging variations in skin texture, lesion shape, depigmentation contrast, lighting condition
Externí odkaz:
http://arxiv.org/abs/2209.01527
Experience replay is an essential component in deep reinforcement learning (DRL), which stores the experiences and generates experiences for the agent to learn in real time. Recently, prioritized experience replay (PER) has been proven to be powerful
Externí odkaz:
http://arxiv.org/abs/2207.07791
Autor:
Mao, Ruibin, Wen, Bo, Zhao, Yahui, Kazemi, Arman, Laguna, Ann Franchesca, Neimier, Michael, Hu, X. Sharon, Sheng, Xia, Graves, Catherine E., Strachan, John Paul, Li, Can
Lifelong on-device learning is a key challenge for machine intelligence, and this requires learning from few, often single, samples. Memory augmented neural network has been proposed to achieve the goal, but the memory module has to be stored in an o
Externí odkaz:
http://arxiv.org/abs/2204.07429
Compression is a standard procedure for making convolutional neural networks (CNNs) adhere to some specific computing resource constraints. However, searching for a compressed architecture typically involves a series of time-consuming training/valida
Externí odkaz:
http://arxiv.org/abs/2107.02927
Autor:
Kazemi, Arman, Sharifi, Mohammad Mehdi, Zou, Zhuowen, Niemier, Michael, Hu, X. Sharon, Imani, Mohsen
Hyperdimensional Computing (HDC) is an emerging computational framework that mimics important brain functions by operating over high-dimensional vectors, called hypervectors (HVs). In-memory computing implementations of HDC are desirable since they c
Externí odkaz:
http://arxiv.org/abs/2106.12029
Autor:
Sharifi, Mohammad Mehdi, Pentecost, Lillian, Rajaei, Ramin, Kazemi, Arman, Lou, Qiuwen, Wei, Gu-Yeon, Brooks, David, Ni, Kai, Hu, X. Sharon, Niemier, Michael, Donato, Marco
The memory wall bottleneck is a key challenge across many data-intensive applications. Multi-level FeFET-based embedded non-volatile memories are a promising solution for denser and more energy-efficient on-chip memory. However, reliable multi-level
Externí odkaz:
http://arxiv.org/abs/2106.11757