no code implementations • 4 Jul 2023 • Kaiqi Zhang, Zixuan Zhang, Minshuo Chen, Yuma Takeda, Mengdi Wang, Tuo Zhao, Yu-Xiang Wang
Convolutional residual neural networks (ConvResNets), though overparameterized, can achieve remarkable prediction performance in practice, which cannot be well explained by conventional wisdom.
no code implementations • 13 Jun 2022 • Kaiqi Zhang, Ming Yin, Yu-Xiang Wang
We propose a quasi neural network to approximate the distribution propagation, which is a neural network with continuous parameters and smooth activation function.
no code implementations • 20 Apr 2022 • Kaiqi Zhang, Yu-Xiang Wang
We consider a "Parallel NN" variant of deep ReLU networks and show that the standard weight decay is equivalent to promoting the $\ell_p$-sparsity ($0<p<1$) of the coefficient vector of an end-to-end learned function bases, i. e., a dictionary.
no code implementations • 11 May 2021 • Yao Chen, Cole Hawkins, Kaiqi Zhang, Zheng Zhang, Cong Hao
This paper emphasizes the importance and efficacy of training, quantization and accelerator design, and calls for more research breakthroughs in the area for AI on the edge.
1 code implementation • 29 Oct 2019 • Chunfeng Cui, Kaiqi Zhang, Talgat Daulbaev, Julia Gusak, Ivan Oseledets, Zheng Zhang
Secondly, we propose analyzing the vulnerability of a neural network using active subspace and finding an additive universal adversarial attack vector that can misclassify a dataset with a high probability.
no code implementations • 28 Jun 2019 • Kaiqi Zhang, Xiyuan Zhang, Zheng Zhang
This paper presents an hardware accelerator for a classical tensor computation framework, Tucker decomposition.
Signal Processing Hardware Architecture
no code implementations • 5 Nov 2018 • Shaokai Ye, Tianyun Zhang, Kaiqi Zhang, Jiayu Li, Jiaming Xie, Yun Liang, Sijia Liu, Xue Lin, Yanzhi Wang
Both DNN weight pruning and clustering/quantization, as well as their combinations, can be solved in a unified manner.
no code implementations • ICLR 2019 • Shaokai Ye, Tianyun Zhang, Kaiqi Zhang, Jiayu Li, Kaidi Xu, Yunfei Yang, Fuxun Yu, Jian Tang, Makan Fardad, Sijia Liu, Xiang Chen, Xue Lin, Yanzhi Wang
Motivated by dynamic programming, the proposed method reaches extremely high pruning rate by using partial prunings with moderate pruning rates.
1 code implementation • 29 Jul 2018 • Tianyun Zhang, Shaokai Ye, Kaiqi Zhang, Xiaolong Ma, Ning Liu, Linfeng Zhang, Jian Tang, Kaisheng Ma, Xue Lin, Makan Fardad, Yanzhi Wang
Without loss of accuracy on the AlexNet model, we achieve 2. 58X and 3. 65X average measured speedup on two GPUs, clearly outperforming the prior work.
3 code implementations • ECCV 2018 • Tianyun Zhang, Shaokai Ye, Kaiqi Zhang, Jian Tang, Wujie Wen, Makan Fardad, Yanzhi Wang
We first formulate the weight pruning problem of DNNs as a nonconvex optimization problem with combinatorial constraints specifying the sparsity requirements, and then adopt the ADMM framework for systematic weight pruning.