Optimization over Continuous and Multi-dimensional Decisions with Observational Data
© 2018 Curran Associates Inc.All rights reserved. We consider the optimization of an uncertain objective over continuous and multidimensional decision spaces in problems in which we are only provided with observational data. We propose a novel algorithmic framework that is tractable, asymptotically...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
2021
|
Online Access: | https://hdl.handle.net/1721.1/137378 |
_version_ | 1810975984845848576 |
---|---|
author | Bertsimas, D McCord, C |
author_facet | Bertsimas, D McCord, C |
author_sort | Bertsimas, D |
collection | MIT |
description | © 2018 Curran Associates Inc.All rights reserved. We consider the optimization of an uncertain objective over continuous and multidimensional decision spaces in problems in which we are only provided with observational data. We propose a novel algorithmic framework that is tractable, asymptotically consistent, and superior to comparable methods on example problems. Our approach leverages predictive machine learning methods and incorporates information on the uncertainty of the predicted outcomes for the purpose of prescribing decisions. We demonstrate the efficacy of our method on examples involving both synthetic and real data sets. |
first_indexed | 2024-09-23T08:48:04Z |
format | Article |
id | mit-1721.1/137378 |
institution | Massachusetts Institute of Technology |
language | English |
last_indexed | 2024-09-23T08:48:04Z |
publishDate | 2021 |
record_format | dspace |
spelling | mit-1721.1/1373782021-11-05T03:38:13Z Optimization over Continuous and Multi-dimensional Decisions with Observational Data Bertsimas, D McCord, C © 2018 Curran Associates Inc.All rights reserved. We consider the optimization of an uncertain objective over continuous and multidimensional decision spaces in problems in which we are only provided with observational data. We propose a novel algorithmic framework that is tractable, asymptotically consistent, and superior to comparable methods on example problems. Our approach leverages predictive machine learning methods and incorporates information on the uncertainty of the predicted outcomes for the purpose of prescribing decisions. We demonstrate the efficacy of our method on examples involving both synthetic and real data sets. 2021-11-04T17:16:47Z 2021-11-04T17:16:47Z 2018-12 2021-02-04T18:39:50Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/137378 Bertsimas, D and McCord, C. 2018. "Optimization over Continuous and Multi-dimensional Decisions with Observational Data." Advances in Neural Information Processing Systems, 2018-December. en Advances in Neural Information Processing Systems Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf Neural Information Processing Systems (NIPS) |
spellingShingle | Bertsimas, D McCord, C Optimization over Continuous and Multi-dimensional Decisions with Observational Data |
title | Optimization over Continuous and Multi-dimensional Decisions with Observational Data |
title_full | Optimization over Continuous and Multi-dimensional Decisions with Observational Data |
title_fullStr | Optimization over Continuous and Multi-dimensional Decisions with Observational Data |
title_full_unstemmed | Optimization over Continuous and Multi-dimensional Decisions with Observational Data |
title_short | Optimization over Continuous and Multi-dimensional Decisions with Observational Data |
title_sort | optimization over continuous and multi dimensional decisions with observational data |
url | https://hdl.handle.net/1721.1/137378 |
work_keys_str_mv | AT bertsimasd optimizationovercontinuousandmultidimensionaldecisionswithobservationaldata AT mccordc optimizationovercontinuousandmultidimensionaldecisionswithobservationaldata |