Conference paper · 2020 · Open access

PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-Time Execution on Mobile Devices

Xiaolong Ma, Fu-Ming Guo, Wei Niu, Xue Lin, Jian Tang, Kaisheng Ma, Bin Ren, Yanzhi Wang

Proceedings of the AAAI Conference on Artificial Intelligence, 34(04), 5117–5124 · Published 3 April 2020 · DOI 10.1609/aaai.v34i04.5954

Open full textCite this paper

In brief

“PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-Time Execution on Mobile Devices” is a 2020 conference paper by Xiaolong Ma and 7 co-authors, published in Proceedings of the AAAI Conference on Artificial Intelligence. OpenAlex counts 233 citations as of 19 November 2026, which puts it in the top 7% of works published in 2020 by citations. It is open access: the full text is free to read at Proceedings of the AAAI Conference on Artificial Intelligence.

Citations
233
Most cited in
2022 (49)
Field-weighted impact
9.25
References
37
Authors
8

Abstract

Model compression techniques on Deep Neural Network (DNN) have been widely acknowledged as an effective way to achieve acceleration on a variety of platforms, and DNN weight pruning is a straightforward and effective method. There are currently two mainstreams of pruning methods representing two extremes of pruning regularity: non-structured, fine-grained pruning can achieve high sparsity and accuracy, but is not hardware friendly; structured, coarse-grained pruning exploits hardware-efficient structures in pruning, but suffers from accuracy drop when the pruning rate is high. In this paper, we introduce PCONV, comprising a new sparsity dimension, – fine-grained pruning patterns inside the coarse-grained structures. PCONV comprises two types of sparsities, Sparse Convolution Patterns (SCP) which is generated from intra-convolution kernel pruning and connectivity sparsity generated from inter-convolution kernel pruning. Essentially, SCP enhances accuracy due to its special vision properties, and connectivity sparsity increases pruning rate while maintaining balanced workload on filter computation. To deploy PCONV, we develop a novel compiler-assisted DNN inference framework and execute PCONV models in real-time without accuracy compromise, which cannot be achieved in prior work. Our experimental results show that, PCONV outperforms three state-of-art end-to-end DNN frameworks, TensorFlow-Lite, TVM, and Alibaba Mobile Neural Network with speedup up to 39.2 ×, 11.4 ×, and 6.3 ×, respectively, with no accuracy loss. Mobile devices can achieve real-time inference on large-scale DNNs.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Ma, X., Guo, F.-M., Niu, W., Lin, X., Tang, J., Ma, K., Ren, B., & Wang, Y. (2020). PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-Time Execution on Mobile Devices. Proceedings of the AAAI Conference on Artificial Intelligence, 34(04), 5117–5124. https://doi.org/10.1609/aaai.v34i04.5954

MLA 9

Ma, Xiaolong, et al. “PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-Time Execution on Mobile Devices.” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, no. 04, 2020, pp. 5117–5124. https://doi.org/10.1609/aaai.v34i04.5954.

Chicago

Ma, Xiaolong, Fu-Ming Guo, Wei Niu, Xue Lin, Jian Tang, Kaisheng Ma, Bin Ren, and Yanzhi Wang. 2020. “PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-Time Execution on Mobile Devices.” Proceedings of the AAAI Conference on Artificial Intelligence 34 (04): 5117–5124. https://doi.org/10.1609/aaai.v34i04.5954.

Harvard

Ma, X. et al. (2020) ‘PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-Time Execution on Mobile Devices’, Proceedings of the AAAI Conference on Artificial Intelligence, 34(04), pp. 5117–5124. doi:10.1609/aaai.v34i04.5954.

IEEE

X. Ma et al., “PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-Time Execution on Mobile Devices,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, no. 04, pp. 5117–5124, 2020, doi: 10.1609/aaai.v34i04.5954.

BibTeX

@inproceedings{ma2020pconv,
  title     = {{PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-Time Execution on Mobile Devices}},
  author    = {Ma, Xiaolong and Guo, Fu-Ming and Niu, Wei and Lin, Xue and Tang, Jian and Ma, Kaisheng and Ren, Bin and Wang, Yanzhi},
  booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year      = {2020},
  volume    = {34},
  number    = {04},
  pages     = {5117--5124},
  doi       = {10.1609/aaai.v34i04.5954},
  url       = {https://doi.org/10.1609/aaai.v34i04.5954}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
201922
20202828
20213737
20224949
20232727
20243434
20253636
20262020

Its field-weighted citation impact is 9.25: it has been cited 9.25 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications

  • model compression
  • real-time inference
  • mobile devices
  • partial convolution

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Ma, X., Guo, F.-M., Niu, W., Lin, X., Tang, J., Ma, K., Ren, B., & Wang, Y. (2020). PCONV: The Missing but Desirable Sparsity in DNN Weight Pruning for Real-Time Execution on Mobile Devices. Proceedings of the AAAI Conference on Artificial Intelligence, 34(04), 5117–5124. https://doi.org/10.1609/aaai.v34i04.5954

How many times has this paper been cited?

233 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at Proceedings of the AAAI Conference on Artificial Intelligence: sign in and open the full text from this page.

Who wrote this paper?

Xiaolong Ma, Fu-Ming Guo, Wei Niu, Xue Lin, Jian Tang, Kaisheng Ma, Bin Ren, Yanzhi Wang.

What is the DOI of this paper?

10.1609/aaai.v34i04.5954, permanently at https://doi.org/10.1609/aaai.v34i04.5954. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.