Numeric Encoding Options with Automunge

Mainstream practice in machine learning with tabular data may take for granted that any feature engineering beyond scaling for numeric sets is superfluous in context of deep neural networks. This paper will offer arguments for potential benefits of extended encodings of numeric streams in deep learn...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Teague, Nicholas J
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Teague, Nicholas J
description Mainstream practice in machine learning with tabular data may take for granted that any feature engineering beyond scaling for numeric sets is superfluous in context of deep neural networks. This paper will offer arguments for potential benefits of extended encodings of numeric streams in deep learning by way of a survey of options for numeric transformations as available in the Automunge open source python library platform for tabular data pipelines, where transformations may be applied to distinct columns in "family tree" sets with generations and branches of derivations. Automunge transformation options include normalization, binning, noise injection, derivatives, and more. The aggregation of these methods into family tree sets of transformations are demonstrated for use to present numeric features to machine learning in multiple configurations of varying information content, as may be applied to encode numeric sets of unknown interpretation. Experiments demonstrate the realization of a novel generalized solution to data augmentation by noise injection for tabular learning, as may materially benefit model performance in applications with underserved training data.
doi_str_mv 10.48550/arxiv.2202.09496
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2202_09496</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2202_09496</sourcerecordid><originalsourceid>FETCH-LOGICAL-a676-99208e518f50a5dfd9be4edd0420292afa821d30137ce1807043741d26be81f73</originalsourceid><addsrcrecordid>eNotzjsPgjAYheEuDkb9AU6yOYFfb7QdDcFLQnRxJ5W22kTAIHj59yo6ne09D0JTDBGTnMNCN09_jwgBEoFiKh6i-a4rbeOLIK2K2vjqFOyvra-rW_Dw7TlYdm1ddtXJjtHA6cvNTv47QodVekg2YbZfb5NlFupYxKFSBKTlWDoOmhtn1NEyawywz6Ui2mlJsKGAqSgsliCAUcGwIfHRSuwEHaHZL9tL82vjS9288q8478X0DXt_OfQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Numeric Encoding Options with Automunge</title><source>arXiv.org</source><creator>Teague, Nicholas J</creator><creatorcontrib>Teague, Nicholas J</creatorcontrib><description>Mainstream practice in machine learning with tabular data may take for granted that any feature engineering beyond scaling for numeric sets is superfluous in context of deep neural networks. This paper will offer arguments for potential benefits of extended encodings of numeric streams in deep learning by way of a survey of options for numeric transformations as available in the Automunge open source python library platform for tabular data pipelines, where transformations may be applied to distinct columns in "family tree" sets with generations and branches of derivations. Automunge transformation options include normalization, binning, noise injection, derivatives, and more. The aggregation of these methods into family tree sets of transformations are demonstrated for use to present numeric features to machine learning in multiple configurations of varying information content, as may be applied to encode numeric sets of unknown interpretation. Experiments demonstrate the realization of a novel generalized solution to data augmentation by noise injection for tabular learning, as may materially benefit model performance in applications with underserved training data.</description><identifier>DOI: 10.48550/arxiv.2202.09496</identifier><language>eng</language><subject>Computer Science - Learning</subject><creationdate>2022-02</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2202.09496$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2202.09496$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Teague, Nicholas J</creatorcontrib><title>Numeric Encoding Options with Automunge</title><description>Mainstream practice in machine learning with tabular data may take for granted that any feature engineering beyond scaling for numeric sets is superfluous in context of deep neural networks. This paper will offer arguments for potential benefits of extended encodings of numeric streams in deep learning by way of a survey of options for numeric transformations as available in the Automunge open source python library platform for tabular data pipelines, where transformations may be applied to distinct columns in "family tree" sets with generations and branches of derivations. Automunge transformation options include normalization, binning, noise injection, derivatives, and more. The aggregation of these methods into family tree sets of transformations are demonstrated for use to present numeric features to machine learning in multiple configurations of varying information content, as may be applied to encode numeric sets of unknown interpretation. Experiments demonstrate the realization of a novel generalized solution to data augmentation by noise injection for tabular learning, as may materially benefit model performance in applications with underserved training data.</description><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzjsPgjAYheEuDkb9AU6yOYFfb7QdDcFLQnRxJ5W22kTAIHj59yo6ne09D0JTDBGTnMNCN09_jwgBEoFiKh6i-a4rbeOLIK2K2vjqFOyvra-rW_Dw7TlYdm1ddtXJjtHA6cvNTv47QodVekg2YbZfb5NlFupYxKFSBKTlWDoOmhtn1NEyawywz6Ui2mlJsKGAqSgsliCAUcGwIfHRSuwEHaHZL9tL82vjS9288q8478X0DXt_OfQ</recordid><startdate>20220218</startdate><enddate>20220218</enddate><creator>Teague, Nicholas J</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20220218</creationdate><title>Numeric Encoding Options with Automunge</title><author>Teague, Nicholas J</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a676-99208e518f50a5dfd9be4edd0420292afa821d30137ce1807043741d26be81f73</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Teague, Nicholas J</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Teague, Nicholas J</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Numeric Encoding Options with Automunge</atitle><date>2022-02-18</date><risdate>2022</risdate><abstract>Mainstream practice in machine learning with tabular data may take for granted that any feature engineering beyond scaling for numeric sets is superfluous in context of deep neural networks. This paper will offer arguments for potential benefits of extended encodings of numeric streams in deep learning by way of a survey of options for numeric transformations as available in the Automunge open source python library platform for tabular data pipelines, where transformations may be applied to distinct columns in "family tree" sets with generations and branches of derivations. Automunge transformation options include normalization, binning, noise injection, derivatives, and more. The aggregation of these methods into family tree sets of transformations are demonstrated for use to present numeric features to machine learning in multiple configurations of varying information content, as may be applied to encode numeric sets of unknown interpretation. Experiments demonstrate the realization of a novel generalized solution to data augmentation by noise injection for tabular learning, as may materially benefit model performance in applications with underserved training data.</abstract><doi>10.48550/arxiv.2202.09496</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2202.09496
ispartof
issn
language eng
recordid cdi_arxiv_primary_2202_09496
source arXiv.org
subjects Computer Science - Learning
title Numeric Encoding Options with Automunge
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T08%3A38%3A18IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Numeric%20Encoding%20Options%20with%20Automunge&rft.au=Teague,%20Nicholas%20J&rft.date=2022-02-18&rft_id=info:doi/10.48550/arxiv.2202.09496&rft_dat=%3Carxiv_GOX%3E2202_09496%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true