2026-09-18T21:25:10Zhttps://dspace.mit.edu/server/oai/request

oai:dspace.mit.edu:1721.1/1215992026-06-06T00:48:40Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023

Randall Davis and Andrew W. Lo. Mishra, Sudhanshu Nath. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science 2019-07-12T17:40:46Z 2019-07-12T17:40:46Z 2018 2018 https://hdl.handle.net/1721.1/121599 1098174801 Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018 Cataloged from PDF version of thesis. Includes bibliographical references (pages 129-131). Deep learning models have demonstrated unprecedented accuracy in wide-ranging tasks such as object and speech recognition. These models can outperform techniques traditionally used in credit risk modeling like logistic regression. However, deep learning models operate as black-boxes, which can limit their use and impact. Regulation mandates that a lender must be able to disclose up to four factors that adversely affected a rejected credit applicant. But we argue that knowing why an applicant is turned down is not enough. An applicant would also want actionable advice that can enable them to reach a favorable classification. Our research thus focuses on both the desire to explain why a machine learning model predicted the classification it did and to find small changes to an input point that can reverse its classification. In this thesis, we evaluate two variants of LIME, a local model-approximation technique and use them in a generate and test algorithm to produce mathematically-effective modifications. We demonstrate that such modifications may not be pragmatically-useful and show how numerical analyses can be supplemented with domain knowledge to generate explanations that are of pragmatic utility. Our work can help accelerate the adoption of deep learning in domains that would benefit from interpreting machine learning predictions. by Sudhanshu Nath Mishra. M.Eng. M.Eng. Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science 131 pages eng Massachusetts Institute of Technology MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission. http://dspace.mit.edu/handle/1721.1/7582 Electrical Engineering and Computer Science. Explaining machine learning predictions : rationales and effective modifications Thesis application/pdf 2019-07-12T17:40:44Z Publication Master EECS unknown unknown <Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="0b23395b-e49b-43ac-b69c-99a69f926a2c"> <Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type> <Language>eng</Language> <Title>Explaining machine learning predictions : rationales and effective modifications</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2018</PublicationDate> <Authors> <Author> <DisplayName>Mishra, Sudhanshu Nath.</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <DisplayName>Massachusetts Institute of Technology</DisplayName> <OrgUnit /> </Publisher> </Publishers> <License>http://dspace.mit.edu/handle/1721.1/7582</License> <Keyword>Electrical Engineering and Computer Science.</Keyword> <Abstract>Deep learning models have demonstrated unprecedented accuracy in wide-ranging tasks such as object and speech recognition. These models can outperform techniques traditionally used in credit risk modeling like logistic regression. However, deep learning models operate as black-boxes, which can limit their use and impact. Regulation mandates that a lender must be able to disclose up to four factors that adversely affected a rejected credit applicant. But we argue that knowing why an applicant is turned down is not enough. An applicant would also want actionable advice that can enable them to reach a favorable classification. Our research thus focuses on both the desire to explain why a machine learning model predicted the classification it did and to find small changes to an input point that can reverse its classification. In this thesis, we evaluate two variants of LIME, a local model-approximation technique and use them in a generate and test algorithm to produce mathematically-effective modifications. We demonstrate that such modifications may not be pragmatically-useful and show how numerical analyses can be supplemented with domain knowledge to generate explanations that are of pragmatic utility. Our work can help accelerate the adoption of deep learning in domains that would benefit from interpreting machine learning predictions.</Abstract> <Access xmlns="http://purl.org/coar/access_right" > </Access> </Publication>