{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/a-practical-tutorial-on-autoencoders-for","title":"A practical tutorial on autoencoders for nonlinear feature fusion: Taxonomy, models, software and guidelines","arxiv_id":"1801.01586","date":"2018-01-04","proceeding":null,"authors":["David Charte","Francisco Charte","Salvador García","María J. del Jesus","Francisco Herrera"],"abstract":"Many of the existing machine learning algorithms, both supervised and\nunsupervised, depend on the quality of the input characteristics to generate a\ngood model. The amount of these variables is also important, since performance\ntends to decline as the input dimensionality increases, hence the interest in\nusing feature fusion techniques, able to produce feature sets that are more\ncompact and higher level. A plethora of procedures to fuse original variables\nfor producing new ones has been developed in the past decades. The most basic\nones use linear combinations of the original variables, such as PCA (Principal\nComponent Analysis) and LDA (Linear Discriminant Analysis), while others find\nmanifold embeddings of lower dimensionality based on non-linear combinations,\nsuch as Isomap or LLE (Linear Locally Embedding) techniques.\n  More recently, autoencoders (AEs) have emerged as an alternative to manifold\nlearning for conducting nonlinear feature fusion. Dozens of AE models have been\nproposed lately, each with its own specific traits. Although many of them can\nbe used to generate reduced feature sets through the fusion of the original\nones, there also AEs designed with other applications in mind.\n  The goal of this paper is to provide the reader with a broad view of what an\nAE is, how they are used for feature fusion, a taxonomy gathering a broad range\nof models, and how they relate to other classical techniques. In addition, a\nset of didactic guidelines on how to choose the proper AE for a given task is\nsupplied, together with a discussion of the software tools available. Finally,\ntwo case studies illustrate the usage of AEs with datasets of handwritten\ndigits and breast cancer.","url_abs":"http://arxiv.org/abs/1801.01586v1","url_pdf":"http://arxiv.org/pdf/1801.01586v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"a-practical-tutorial-on-autoencoders-for","repo_url":"https://github.com/fdavidcl/ae-review-resources","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[],"methods":[{"method_slug":"ae","method_name":"AE"},{"method_slug":"lda","method_name":"LDA"},{"method_slug":"pca","method_name":"PCA"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}