Unsupervised pre-training for fully convolutional neural networks
| dc.contributor.author | Wiehman, Stiaan | en_ZA |
| dc.contributor.author | Kroon, Steve | en_ZA |
| dc.contributor.author | De Villiers, Hendrik | en_ZA |
| dc.date.accessioned | 2017-01-23T13:50:17Z | |
| dc.date.available | 2017-01-23T13:50:17Z | |
| dc.date.issued | 2016 | |
| dc.description | CITATION: Wiehman, S., Kroon, S. & De Villiers, H. 2016. Unsupervised pre-training for fully convolutional neural networks. Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 30 November-2 December 2016, Stellenbosch, South Africa. | |
| dc.description | The original publication is available at http://ieeexplore.ieee.org | |
| dc.description.abstract | Unsupervised pre-training of neural networks has been shown to act as a regularization technique, improving performance and reducing model variance. Recently, fully con-volutional networks (FCNs) have shown state-of-the-art results on various semantic segmentation tasks. Unfortunately, there is no efficient approach available for FCNs to benefit from unsupervised pre-training. Given the unique property of FCNs to output segmentation maps, we explore a novel variation of unsupervised pre-training specifically designed for FCNs. We extend an existing FCN, called U-net, to facilitate end-to-end unsupervised pre-training and apply it on the ISBI 2012 EM segmentation challenge data set. We performed a battery of significance tests for both equality of means and equality of variance, and show that our results are consistent with previous work on unsupervised pre-training obtained from much smaller networks. We conclude that end-to-end unsupervised pre-training for FCNs adds robustness to random initialization, thus reducing model variance. | en_ZA |
| dc.description.uri | http://ieeexplore.ieee.org/document/7813160/ | |
| dc.description.version | Post print | |
| dc.format.extent | 6 pages ; illustrations | en_ZA |
| dc.identifier.citation | Wiehman, S., Kroon, S. & De Villiers, H. 2016. Unsupervised pre-training for fully convolutional neural networks. Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 30 November-2 December 2016, Stellenbosch, South Africa | |
| dc.identifier.isbn | 978-1-5090-3335-5 (online) | |
| dc.identifier.isbn | 978-1-5090-3336-2 (print) | |
| dc.identifier.other | doi:10.1109/RoboMech.2016.7813160 | |
| dc.identifier.uri | http://hdl.handle.net/10019.1/100503 | |
| dc.language.iso | en_ZA | en_ZA |
| dc.publisher | Institute of Electrical and Electronics Engineers | en_ZA |
| dc.rights.holder | Institute of Electrical and Electronics Engineers | en_ZA |
| dc.subject | Neural networks (Computer science) | en_ZA |
| dc.subject | Convolutions (Mathematics) | en_ZA |
| dc.subject | Map segmentation -- Semantics | en_ZA |
| dc.title | Unsupervised pre-training for fully convolutional neural networks | en_ZA |
| dc.type | Conference Paper | en_ZA |