Deep feature learning for dummies: A simple auto-encoder training method using Particle Swarm Optimisation. Sui, C., Bennamoun, M., & Togneri, R. Pattern Recognition Letters, 94:75--80, July, 2017. ![Deep feature learning for dummies: A simple auto-encoder training method using Particle Swarm Optimisation [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex One highlight of deep networks is their ability to automatically learn useful representations from raw inputs. Hence, domain knowledge, generic priors and feature extraction are no longer needed. It is however still a challenging task to train deep networks, because they require extensive human expertise to choose the type of model and its parameters to ensure that the system is properly trained. In this work, we propose a novel feature learning framework based on the marginalised stacked auto-encoder which does not need practitioners to have any deep learning specific knowledge. We applied this method on visual speech recognition, and the performance of our proposed method outperforms the other feature extraction methods with a 2% improvement in the accuracy for speaker independent systems. This method is also a universal solution which can be used for any deep learning based tasks. Therefore, we also verified our method on a popular hand written digit recognition database MNIST, and experimental results showed that our proposed method with an error rate of 1.30% is comparable to the best models tuned by experts.
@article{sui_deep_2017,
title = {Deep feature learning for dummies: {A} simple auto-encoder training method using {Particle} {Swarm} {Optimisation}},
volume = {94},
issn = {0167-8655},
shorttitle = {Deep feature learning for dummies},
url = {http://www.sciencedirect.com/science/article/pii/S0167865517300910},
doi = {10.1016/j.patrec.2017.03.021},
abstract = {One highlight of deep networks is their ability to automatically learn useful representations from raw inputs. Hence, domain knowledge, generic priors and feature extraction are no longer needed. It is however still a challenging task to train deep networks, because they require extensive human expertise to choose the type of model and its parameters to ensure that the system is properly trained. In this work, we propose a novel feature learning framework based on the marginalised stacked auto-encoder which does not need practitioners to have any deep learning specific knowledge. We applied this method on visual speech recognition, and the performance of our proposed method outperforms the other feature extraction methods with a 2\% improvement in the accuracy for speaker independent systems. This method is also a universal solution which can be used for any deep learning based tasks. Therefore, we also verified our method on a popular hand written digit recognition database MNIST, and experimental results showed that our proposed method with an error rate of 1.30\% is comparable to the best models tuned by experts.},
urldate = {2018-03-25TZ},
journal = {Pattern Recognition Letters},
author = {Sui, Chao and Bennamoun, Mohammed and Togneri, Roberto},
month = jul,
year = {2017},
keywords = {Automatic feature learning, Marginalised stacked denoising auto-encoder, Particle Swarm Optimisation, Simple auto-encoder training method, Visual speech recognition},
pages = {75--80}
}
Downloads: 0
{"_id":"pRRxCGPnY6cusSWHM","bibbaseid":"sui-bennamoun-togneri-deepfeaturelearningfordummiesasimpleautoencodertrainingmethodusingparticleswarmoptimisation-2017","downloads":0,"creationDate":"2018-04-15T14:41:43.868Z","title":"Deep feature learning for dummies: A simple auto-encoder training method using Particle Swarm Optimisation","author_short":["Sui, C.","Bennamoun, M.","Togneri, R."],"year":2017,"bibtype":"article","biburl":"https://bibbase.org/zotero/pvhuy","bibdata":{"bibtype":"article","type":"article","title":"Deep feature learning for dummies: A simple auto-encoder training method using Particle Swarm Optimisation","volume":"94","issn":"0167-8655","shorttitle":"Deep feature learning for dummies","url":"http://www.sciencedirect.com/science/article/pii/S0167865517300910","doi":"10.1016/j.patrec.2017.03.021","abstract":"One highlight of deep networks is their ability to automatically learn useful representations from raw inputs. Hence, domain knowledge, generic priors and feature extraction are no longer needed. It is however still a challenging task to train deep networks, because they require extensive human expertise to choose the type of model and its parameters to ensure that the system is properly trained. In this work, we propose a novel feature learning framework based on the marginalised stacked auto-encoder which does not need practitioners to have any deep learning specific knowledge. We applied this method on visual speech recognition, and the performance of our proposed method outperforms the other feature extraction methods with a 2% improvement in the accuracy for speaker independent systems. This method is also a universal solution which can be used for any deep learning based tasks. Therefore, we also verified our method on a popular hand written digit recognition database MNIST, and experimental results showed that our proposed method with an error rate of 1.30% is comparable to the best models tuned by experts.","urldate":"2018-03-25TZ","journal":"Pattern Recognition Letters","author":[{"propositions":[],"lastnames":["Sui"],"firstnames":["Chao"],"suffixes":[]},{"propositions":[],"lastnames":["Bennamoun"],"firstnames":["Mohammed"],"suffixes":[]},{"propositions":[],"lastnames":["Togneri"],"firstnames":["Roberto"],"suffixes":[]}],"month":"July","year":"2017","keywords":"Automatic feature learning, Marginalised stacked denoising auto-encoder, Particle Swarm Optimisation, Simple auto-encoder training method, Visual speech recognition","pages":"75--80","bibtex":"@article{sui_deep_2017,\n\ttitle = {Deep feature learning for dummies: {A} simple auto-encoder training method using {Particle} {Swarm} {Optimisation}},\n\tvolume = {94},\n\tissn = {0167-8655},\n\tshorttitle = {Deep feature learning for dummies},\n\turl = {http://www.sciencedirect.com/science/article/pii/S0167865517300910},\n\tdoi = {10.1016/j.patrec.2017.03.021},\n\tabstract = {One highlight of deep networks is their ability to automatically learn useful representations from raw inputs. Hence, domain knowledge, generic priors and feature extraction are no longer needed. It is however still a challenging task to train deep networks, because they require extensive human expertise to choose the type of model and its parameters to ensure that the system is properly trained. In this work, we propose a novel feature learning framework based on the marginalised stacked auto-encoder which does not need practitioners to have any deep learning specific knowledge. We applied this method on visual speech recognition, and the performance of our proposed method outperforms the other feature extraction methods with a 2\\% improvement in the accuracy for speaker independent systems. This method is also a universal solution which can be used for any deep learning based tasks. Therefore, we also verified our method on a popular hand written digit recognition database MNIST, and experimental results showed that our proposed method with an error rate of 1.30\\% is comparable to the best models tuned by experts.},\n\turldate = {2018-03-25TZ},\n\tjournal = {Pattern Recognition Letters},\n\tauthor = {Sui, Chao and Bennamoun, Mohammed and Togneri, Roberto},\n\tmonth = jul,\n\tyear = {2017},\n\tkeywords = {Automatic feature learning, Marginalised stacked denoising auto-encoder, Particle Swarm Optimisation, Simple auto-encoder training method, Visual speech recognition},\n\tpages = {75--80}\n}\n\n","author_short":["Sui, C.","Bennamoun, M.","Togneri, R."],"key":"sui_deep_2017","id":"sui_deep_2017","bibbaseid":"sui-bennamoun-togneri-deepfeaturelearningfordummiesasimpleautoencodertrainingmethodusingparticleswarmoptimisation-2017","role":"author","urls":{"Paper":"http://www.sciencedirect.com/science/article/pii/S0167865517300910"},"keyword":["Automatic feature learning","Marginalised stacked denoising auto-encoder","Particle Swarm Optimisation","Simple auto-encoder training method","Visual speech recognition"],"downloads":0},"search_terms":["deep","feature","learning","dummies","simple","auto","encoder","training","method","using","particle","swarm","optimisation","sui","bennamoun","togneri"],"keywords":["automatic feature learning","marginalised stacked denoising auto-encoder","particle swarm optimisation","simple auto-encoder training method","visual speech recognition"],"authorIDs":[],"dataSources":["E2tDnyqghkeNYrDGK"]}