Weiwei Chen

Weiwei Chen

Auto-generated profile, based on publicly available information

Weiwei Chen

Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 October 2022 in ACM Transactions on Embedded Computing Systems
Verified
Journal Article
Open access
WCWeiwei Chen
Weiwei Chen
YWYing Wang
Ying Wang
YXYing Xu
Ying Xu
CGChengsi Gao
Chengsi Gao
CLCheng Liu
Cheng Liu
LZLei Zhang
Lei Zhang
The efficiency of deep neural network (DNN) solutions on real hardware devices are mainly decided by the DNN architecture and the compiler-level scheduling strategy on the hardware. When we try to fully exploit the underlying hardware and obtain the optimal tradeoff between DNN accuracy and runtime performance, we discovered that the two optimization goals of DNN architecture and scheduling policy are intimately related to each other. However, current hardware-aware Neural Architecture Search (NAS) methods primarily focus on the DNN architecture search process, ignoring the effects of various compiler-level scheduling strategies (e.g., graph-level optimization, loop transformations, parallelization, etc.) on network candidates being evaluated in the search process. As a result, they may overlook the true-optimal DNN implementations on hardware, which can only be discovered by trying-out different combinations of scheduling strategies and DNN architectures. This work proposes a NAS framework (CHaNAS) that searches for not only the network architecture but also the dedicated compiler-level scheduling policy, as the optimal co-design solution on the target hardware. We propose to use a block-based pre-scheduling methodology to reduce the co-design search space and enable the automatic generation of the optimal co-design, including the network architecture and the tensor programs that practice the scheduling policy. Further, we introduce a new search objective function based on the generalization gap to prevent the selection of architectures that are prone to overfitting. We evaluate CHaNAS on Imagenet on different hardware back-ends against the state-of-the-art hardware-aware search method based on the MobileNet-v3 search space. Experimental results show that the co-design solutions obtained by ChaNAS show up to 1.6×, 1.9×, and 1.7×, 24 performance boost on NVIDIA P100 GPU, Intel Xeon 8163 CPU, and Samsung Note 10 Mobile, respectively, over the baselines of the same-level accuracy.

3 March 2022 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
CGChengsi Gao
Chengsi Gao
YWYing Wang
Ying Wang
YHYinhe Han
Yinhe Han
WCWeiwei Chen
Weiwei Chen
LZLei Zhang
Lei Zhang
Recently, video processing tasks, such as video enhancement and analysis, have received increasing attention from both academics and industries. However, the current video processing procedure on edge decouples the decoding phase and the subsequent video processing tasks, missing the opportunity to accelerate the procedure by orchestrating video decoding and enhancement stages. Thus, we propose an intelligent video processing workflow and architecture(IVP) for cloud-edge video streaming. For edge devices that receive compressed videos, IVP can perform direct DNN-based video enhancement, e.g., super-resolution and frame-interpolation. By leveraging the metadata motion vectors and residuals extracted from the encoded video, our architecture will significantly eliminate unnecessary frame pixels being processed by the DNNs and improve execution efficiency. The proposed IVP and workflow are proved to reduce up to 90% of the processing latency while producing accurate and high-quality videos. Furthermore, we observe a significant portion of similar optical flow in time domain of continuous videos, which can be used to reduce the computation overhead. Thus, to utilize such temporal similarity of optical flow, the proposed IVP is upgraded to be capable of reusing previous computation results, which further improves energy efficiency of the whole system.

6 June 2021
Verified
Conference Paper
Closed access
WCWeiwei Chen
Weiwei Chen
CWChong Wang
Chong Wang
ZZZhehao Zhang
Zhehao Zhang
ZHZheng Huo
Zheng Huo
LGLinlin Gao
Linlin Gao
Computational efficiency of modern deep convolutional networks is always desired in real world applications. Various types of group convolutions have been widely used to reduce the complexity of the networks. Inspired by the previous work, a new reweighted dynamic group convolution (RDGC) structure, including a reweighted pruning module and a survival loss, is proposed in this work for more precise channel pruning. Specifically, a layer-wise pruning rate adjustment strategy is employed with an intra-cluster loss to guide the learning of self-attention implicitly. The proposed model is more capable to select the suitable channels to construct the filter groups. Its effectiveness is demonstrated by extensive experiments on four public datasets, namely CIFAR-10, CIFAR-100 and two fine-grained image datasets (Stanford Dogs and Stanford Cars).

1 June 2020
Verified
Conference Paper
Closed access
WCWeiwei Chen
Weiwei Chen
ZLZhigang Li
Zhigang Li
YTYanqin Tang
Yanqin Tang
BWBaohua Wang
Baohua Wang
BWBin Wang
Bin Wang
YLYing Liu
Ying Liu
The problem of Attention Deficit Disorder is more noticeable in online teaching than in face-to-face teaching. Facing the new requests of constructing Golden Course and based on the analysis of the relationship between Golden Course and Attention Deficit Disorder, the paper takes the national-level MOOC, Data structure, as an example and explores three approaches to solve the problem of Attention Deficit Disorder from the perspective of teaching content organization and implementation. These three approaches are also the core strategies for Golden Course design. The main content is as follows. First, establish the levels of learning objectives under the guidance of BLOOM model, and optimize the design of learning objectives with Mager method, so that students can obtain specific information, find correct learning direction, and avoid lack of attention. Second, use concept map to select and divide learning content, highlighting the importance and relevance of learning content, and ensure that the content is explained within about 10 minutes. Although the time is fragmented, the content is not fragmented. Third, the BOPPPS model is used to organize the process of course implementation and introduce the learning content under the guidance of questions, so as to ensure that the internal logic of the course is compact and effective, and students' interest in learning can be enhanced. Practice shows that the curriculum design strategy which combines the three approaches mentioned above can effectively solve the problem of Attention Deficit Disorder. The course, Data Structure, constructed with three mentioned approaches, was selected as one of the national-level MOOCs in January 2019, which has been running for 4 years on the net for 7 semesters, with a course rating score of 4.8 points out of 5 points, and taken by more than 70,000 learners.

5 December 2023 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
LLLian Liu
Lian Liu
YWYing Wang
Ying Wang
XZXiandong Zhao
Xiandong Zhao
WCWeiwei Chen
Weiwei Chen
HLHuawei Li
Huawei Li
YHYinhe Han
Yinhe Han
Efficient deep learning models, especially optimized for edge devices, benefit from low inference latency to efficient energy consumption. Two classical techniques for efficient model inference are lightweight neural architecture search (NAS), which automatically designs compact network models, and quantization, which reduces the bit-precision of neural network models. As a consequence, joint design for both neural architecture and quantization precision settings is becoming increasingly popular. There are three main aspects that affect the performance of the joint optimization between neural architecture and quantization: 1) quantization precision selection (QPS); 2) quantization-aware training (QAT); and 3) NAS. However, existing works focus on at most twofold of these aspects, and result in secondary performance. To this end, we proposed a novel automatic optimization framework, DAQU, that allows jointly searching for Pareto-optimal neural architecture and quantization precision combination among more than $10^{47}$ quantized subnet models. To overcome the instability of the conventional automatic optimization framework, DAQU incorporates a warm-up strategy to reduce the accuracy gap among different neural architectures, and a precision-transfer training approach to maintain flexibility among different quantization precision settings. Our experiments show that the quantized lightweight neural networks generated by DAQU consistently outperform state-of-the-art NAS and quantization joint optimization methods.

20 October 2022 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
WCWeiwei Chen
Weiwei Chen
YWYing Wang
Ying Wang
YXYing Xu
Ying Xu
CGChengsi Gao
Chengsi Gao
YHYinhe Han
Yinhe Han
LZLei Zhang
Lei Zhang
Dynamic resources management in reconfigurable processors often manifests as a hard online decision-making task, which should yield premier solutions that must meet Quality-of-Service (QoS) requirements while maximizing the system’s efficiency. Most prior works rely on a hard-to-train predictor to model the complicated relationships between processor configurations and performance. To decide the proper resource allocation, the predictor needs to tentatively evaluate a group of possible configurations, and then decide the best configuration for the workload. This tedious process has an expensive runtime overhead for resource configuration in processors. Besides, prior works focus on improving the prediction accuracy, however, higher performance prediction cannot guarantee a good system outcome. Inspired by recent advances in adversarial learning, we present a generative adversarial network (GAN)-based framework, Amphis, which can directly generate the on-demand processor configuration for any scheduled-in application. By evaluating Amphis on a reconfigurable processor with 18 different workloads, our results demonstrate that the GAN-based method provides tremendous overhead reduction (up to 90%) compared to the SOTA prediction-based method WNNM while providing higher resource utilization.