Achievable Minimally-Contrastive Counterfactual Explanations
Decision support systems based on machine learning models should be able to help users identify opportunities and threats. Popular model-agnostic explanation models can identify factors that support various predictions, answering questions such as “What factors affect sales?” or “Why did sales decli...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Machine Learning and Knowledge Extraction |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-4990/5/3/48 |
| _version_ | 1797579072672366592 |
|---|---|
| author | Hosein Barzekar Susan McRoy |
| author_facet | Hosein Barzekar Susan McRoy |
| author_sort | Hosein Barzekar |
| collection | DOAJ |
| description | Decision support systems based on machine learning models should be able to help users identify opportunities and threats. Popular model-agnostic explanation models can identify factors that support various predictions, answering questions such as “What factors affect sales?” or “Why did sales decline?”, but do not highlight what a person should or could do to get a more desirable outcome. Counterfactual explanation approaches address intervention, and some even consider feasibility, but none consider their suitability for real-time applications, such as question answering. Here, we address this gap by introducing a novel model-agnostic method that provides specific, feasible changes that would impact the outcomes of a complex Black Box AI model for a given instance and assess its real-world utility by measuring its real-time performance and ability to find achievable changes. The method uses the instance of concern to generate high-precision explanations and then applies a secondary method to find achievable minimally-contrastive counterfactual explanations (AMCC) while limiting the search to modifications that satisfy domain-specific constraints. Using a widely recognized dataset, we evaluated the classification task to ascertain the frequency and time required to identify successful counterfactuals. For a 90% accurate classifier, our algorithm identified AMCC explanations in 47% of cases (38 of 81), with an average discovery time of 80 ms. These findings verify the algorithm’s efficiency in swiftly producing AMCC explanations, suitable for real-time systems. The AMCC method enhances the transparency of Black Box AI models, aiding individuals in evaluating remedial strategies or assessing potential outcomes. |
| first_indexed | 2024-03-10T22:31:47Z |
| format | Article |
| id | doaj.art-273cd0e28cd445f0be12e92f597a1d98 |
| institution | Directory Open Access Journal |
| issn | 2504-4990 |
| language | English |
| last_indexed | 2024-03-10T22:31:47Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Machine Learning and Knowledge Extraction |
| spelling | doaj.art-273cd0e28cd445f0be12e92f597a1d982023-11-19T11:41:41ZengMDPI AGMachine Learning and Knowledge Extraction2504-49902023-08-015392293610.3390/make5030048Achievable Minimally-Contrastive Counterfactual ExplanationsHosein Barzekar0Susan McRoy1Department of Computer Science, University of Wisconsin-Milwaukee, Milwaukee, WI 53211, USADepartment of Computer Science, University of Wisconsin-Milwaukee, Milwaukee, WI 53211, USADecision support systems based on machine learning models should be able to help users identify opportunities and threats. Popular model-agnostic explanation models can identify factors that support various predictions, answering questions such as “What factors affect sales?” or “Why did sales decline?”, but do not highlight what a person should or could do to get a more desirable outcome. Counterfactual explanation approaches address intervention, and some even consider feasibility, but none consider their suitability for real-time applications, such as question answering. Here, we address this gap by introducing a novel model-agnostic method that provides specific, feasible changes that would impact the outcomes of a complex Black Box AI model for a given instance and assess its real-world utility by measuring its real-time performance and ability to find achievable changes. The method uses the instance of concern to generate high-precision explanations and then applies a secondary method to find achievable minimally-contrastive counterfactual explanations (AMCC) while limiting the search to modifications that satisfy domain-specific constraints. Using a widely recognized dataset, we evaluated the classification task to ascertain the frequency and time required to identify successful counterfactuals. For a 90% accurate classifier, our algorithm identified AMCC explanations in 47% of cases (38 of 81), with an average discovery time of 80 ms. These findings verify the algorithm’s efficiency in swiftly producing AMCC explanations, suitable for real-time systems. The AMCC method enhances the transparency of Black Box AI models, aiding individuals in evaluating remedial strategies or assessing potential outcomes.https://www.mdpi.com/2504-4990/5/3/48machine learninginterpretabilityfeasibilitycounterfactual and contrastive explanation |
| spellingShingle | Hosein Barzekar Susan McRoy Achievable Minimally-Contrastive Counterfactual Explanations Machine Learning and Knowledge Extraction machine learning interpretability feasibility counterfactual and contrastive explanation |
| title | Achievable Minimally-Contrastive Counterfactual Explanations |
| title_full | Achievable Minimally-Contrastive Counterfactual Explanations |
| title_fullStr | Achievable Minimally-Contrastive Counterfactual Explanations |
| title_full_unstemmed | Achievable Minimally-Contrastive Counterfactual Explanations |
| title_short | Achievable Minimally-Contrastive Counterfactual Explanations |
| title_sort | achievable minimally contrastive counterfactual explanations |
| topic | machine learning interpretability feasibility counterfactual and contrastive explanation |
| url | https://www.mdpi.com/2504-4990/5/3/48 |
| work_keys_str_mv | AT hoseinbarzekar achievableminimallycontrastivecounterfactualexplanations AT susanmcroy achievableminimallycontrastivecounterfactualexplanations |