no code implementations • 18 Jan 2024 • Yang Sui, Miao Yin, Yu Gong, Jinqi Xiao, Huy Phan, Bo Yuan
Low-rank compression, a popular model compression technique that produces compact convolutional neural networks (CNNs) with low rankness, has been well-studied in the literature.
1 code implementation • 26 May 2023 • Jinqi Xiao, Miao Yin, Yu Gong, Xiao Zang, Jian Ren, Bo Yuan
Attention-based vision models, such as Vision Transformer (ViT) and its variants, have shown promising performance in various computer vision tasks.
no code implementations • 20 Jan 2023 • Jinqi Xiao, Chengming Zhang, Yu Gong, Miao Yin, Yang Sui, Lizhi Xiang, Dingwen Tao, Bo Yuan
By interpreting automatic rank selection from an architecture search perspective, we develop an end-to-end solution to determine the suitable layer-wise ranks in a differentiable and hardware-aware way.
no code implementations • 13 Jan 2023 • Miao Yin, Burak Uzkent, Yilin Shen, Hongxia Jin, Bo Yuan
We first develop a graph-based ranking for measuring the importance of attention heads, and the extracted importance information is further integrated to an optimization-based procedure to impose the heterogeneous structured sparsity patterns on the ViT models.
no code implementations • 5 Dec 2022 • Yu Gong, Miao Yin, Lingyi Huang, Chunhua Deng, Yang Sui, Bo Yuan
Meanwhile, compared with the state-of-the-art tensor decomposed model-oriented hardware TIE, our proposed FDHT-LSTM architecture achieves better performance in throughput, area efficiency and energy efficiency, respectively on LSTM-Youtube workload.
no code implementations • 4 Dec 2022 • Huy Phan, Miao Yin, Yang Sui, Bo Yuan, Saman Zonouz
Considering the co-importance of model compactness and robustness in practical applications, several prior works have explored to improve the adversarial robustness of the sparse neural networks.
no code implementations • 24 Aug 2022 • Xiao Zang, Miao Yin, Lingyi Huang, Jingjin Yu, Saman Zonouz, Bo Yuan
Despite the current development in this direction, the efficient capture and processing of important sequential and spatial information, in a direct and simultaneous way, is still relatively under-explored.
no code implementations • CVPR 2022 • Miao Yin, Yang Sui, Wanzhao Yang, Xiao Zang, Yu Gong, Bo Yuan
High-order decomposition is a widely used model compression approach towards compact convolutional neural networks (CNNs).
1 code implementation • NeurIPS 2021 • Yang Sui, Miao Yin, Yi Xie, Huy Phan, Saman Zonouz, Bo Yuan
Filter pruning has been widely used for neural network compression because of its enabled practical acceleration.
no code implementations • 29 Sep 2021 • Wanzhao Yang, Miao Yin, Yang Sui, Bo Yuan
Based on the observations and outcomes from our analysis, we then propose SPARK, a unified DNN compression framework that can simultaneously capture model SPArsity and low-RanKness in an efficient way.
no code implementations • CVPR 2021 • Miao Yin, Yang Sui, Siyu Liao, Bo Yuan
Notably, on CIFAR-100, with 2. 3X and 2. 4X compression ratios, our models have 1. 96% and 2. 21% higher top-1 accuracy than the original ResNet-20 and ResNet-32, respectively.
no code implementations • CVPR 2021 • Miao Yin, Siyu Liao, Xiao-Yang Liu, Xiaodong Wang, Bo Yuan
Although various prior works have been proposed to reduce the RNN model sizes, executing RNN models in resource-restricted environments is still a very challenging problem.
no code implementations • 8 Feb 2021 • Siyu Liao, Chunhua Deng, Miao Yin, Bo Yuan
Recently deep neural networks have been successfully applied in channel coding to improve the decoding performance.
no code implementations • 9 May 2020 • Miao Yin, Siyu Liao, Xiao-Yang Liu, Xiaodong Wang, Bo Yuan
Recurrent Neural Networks (RNNs) have been widely used in sequence analysis and modeling.
1 code implementation • 28 Jan 2019 • Zihan Ding, Xiao-Yang Liu, Miao Yin, Linghe Kong
Secondly, we propose TGAN that integrates deep convolutional generative adversarial networks and tensor super-resolution in a cascading manner, to generate high-quality images from random distributions.
no code implementations • 8 Apr 2017 • Feng Qian, Miao Yin, Ming-Jun Su, Yaojun Wang, Guangmin Hu
Prestack seismic data carries much useful information that can help us find more complex atypical reservoirs.