Preprint · 2015 · Open access

Deep Clustered Convolutional Kernels

Minyoung Kim, Luca Rigazio

arXiv (Cornell University), 160–172 · Published 6 March 2015 · DOI 10.48550/arxiv.1503.01824

Open PDFView at the publisherCite this paper

In brief

“Deep Clustered Convolutional Kernels” is a 2015 preprint by Minyoung Kim and Luca Rigazio, published in arXiv (Cornell University). OpenAlex counts 16 citations as of 18 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
16
Most cited in
2018 (5)
References
32
Authors
2

Abstract

Deep neural networks have recently achieved state of the art performance thanks to new training algorithms for rapid parameter estimation and new regularization methods to reduce overfitting. However, in practice the network architecture has to be manually set by domain experts, generally by a costly trial and error procedure, which often accounts for a large portion of the final system performance. We view this as a limitation and propose a novel training algorithm that automatically optimizes network architecture, by progressively increasing model complexity and then eliminating model redundancy by selectively removing parameters at training time. For convolutional neural networks, our method relies on iterative split/merge clustering of convolutional kernels interleaved by stochastic gradient descent. We present a training algorithm and experimental results on three different vision tasks, showing improved performance compared to similarly sized hand-crafted architectures.

Cite this paper

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

APA 7

Kim, M., & Rigazio, L. (2015). Deep Clustered Convolutional Kernels. arXiv (Cornell University), 160–172. https://doi.org/10.48550/arxiv.1503.01824

MLA 9

Kim, Minyoung, and Luca Rigazio. “Deep Clustered Convolutional Kernels.” arXiv (Cornell University), 2015, pp. 160–172. https://doi.org/10.48550/arxiv.1503.01824.

Chicago

Kim, Minyoung and Luca Rigazio. 2015. “Deep Clustered Convolutional Kernels.” arXiv (Cornell University): 160–172. https://doi.org/10.48550/arxiv.1503.01824.

Harvard

Kim, M. and Rigazio, L. (2015) ‘Deep Clustered Convolutional Kernels’, arXiv (Cornell University), pp. 160–172. doi:10.48550/arxiv.1503.01824.

IEEE

M. Kim and L. Rigazio, “Deep Clustered Convolutional Kernels,” arXiv (Cornell University), pp. 160–172, 2015, doi: 10.48550/arxiv.1503.01824.

BibTeX

@misc{kim2015deep,
  title     = {{Deep Clustered Convolutional Kernels}},
  author    = {Kim, Minyoung and Rigazio, Luca},
  howpublished = {arXiv (Cornell University)},
  year      = {2015},
  pages     = {160--172},
  doi       = {10.48550/arxiv.1503.01824},
  url       = {https://doi.org/10.48550/arxiv.1503.01824}
}

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

Citations by year

YearCitationsShare of the best year
201511
201711
201855
201911
202044
202122
202222

Topic and keywords

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

  • convolutional neural networks
  • kernel clustering
  • parameter pruning
  • visual task

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?

Kim, M., & Rigazio, L. (2015). Deep Clustered Convolutional Kernels. arXiv (Cornell University), 160–172. https://doi.org/10.48550/arxiv.1503.01824

How many times has this paper been cited?

16 times, according to OpenAlex on 18 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 arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Minyoung Kim, Luca Rigazio.

What is the DOI of this paper?

10.48550/arxiv.1503.01824, permanently at https://doi.org/10.48550/arxiv.1503.01824. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 18 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.