Matches in SemOpenAlex for { <https://semopenalex.org/work/W3035625410> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W3035625410 abstract "Model interpretation is essential in data mining and knowledge discovery. It can help understand the intrinsic model working mechanism and check if the model has undesired characteristics. A popular way of performing model interpretation is Instance-wise Feature Selection (IFS), which provides an importance score of each feature representing the data samples to explain how the model generates the specific output. In this paper, we propose a Model-agnostic Effective Efficient Direct (MEED) IFS framework for model interpretation, mitigating concerns about sanity, combinatorial shortcuts, model identifiability, and information transmission. Also, we focus on the following setting: using selected features to directly predict the output of the given model, which serves as a primary evaluation metric for model-interpretation methods. Apart from the features, we involve the output of the given model as an additional input to learn an explainer based on more accurate information. To learn the explainer, besides fidelity, we propose an Adversarial Infidelity Learning (AIL) mechanism to boost the explanation learning by screening relatively unimportant features. Through theoretical and experimental analysis, we show that our AIL mechanism can help learn the desired conditional distribution between selected features and targets. Moreover, we extend our framework by integrating efficient interpretation methods as proper priors to provide a warm start. Comprehensive empirical evaluation results are provided by quantitative metrics and human evaluation to demonstrate the effectiveness and superiority of our proposed method. Our code is publicly available online at https://github.com/langlrsw/MEED." @default.
- W3035625410 created "2020-06-19" @default.
- W3035625410 creator A5031863894 @default.
- W3035625410 creator A5081211506 @default.
- W3035625410 creator A5086529957 @default.
- W3035625410 creator A5086988988 @default.
- W3035625410 creator A5090022501 @default.
- W3035625410 date "2020-08-20" @default.
- W3035625410 modified "2023-10-17" @default.
- W3035625410 title "Adversarial Infidelity Learning for Model Interpretation" @default.
- W3035625410 cites W1787224781 @default.
- W3035625410 cites W1901129140 @default.
- W3035625410 cites W2108598243 @default.
- W3035625410 cites W2112796928 @default.
- W3035625410 cites W2282821441 @default.
- W3035625410 cites W2952622013 @default.
- W3035625410 cites W2964296416 @default.
- W3035625410 cites W2970726176 @default.
- W3035625410 cites W2980096013 @default.
- W3035625410 cites W2996061341 @default.
- W3035625410 cites W3138819813 @default.
- W3035625410 doi "https://doi.org/10.1145/3394486.3403071" @default.
- W3035625410 hasPublicationYear "2020" @default.
- W3035625410 type Work @default.
- W3035625410 sameAs 3035625410 @default.
- W3035625410 citedByCount "9" @default.
- W3035625410 countsByYear W30356254102020 @default.
- W3035625410 countsByYear W30356254102021 @default.
- W3035625410 countsByYear W30356254102022 @default.
- W3035625410 countsByYear W30356254102023 @default.
- W3035625410 crossrefType "proceedings-article" @default.
- W3035625410 hasAuthorship W3035625410A5031863894 @default.
- W3035625410 hasAuthorship W3035625410A5081211506 @default.
- W3035625410 hasAuthorship W3035625410A5086529957 @default.
- W3035625410 hasAuthorship W3035625410A5086988988 @default.
- W3035625410 hasAuthorship W3035625410A5090022501 @default.
- W3035625410 hasBestOaLocation W30356254102 @default.
- W3035625410 hasConcept C107673813 @default.
- W3035625410 hasConcept C111919701 @default.
- W3035625410 hasConcept C119857082 @default.
- W3035625410 hasConcept C122770356 @default.
- W3035625410 hasConcept C124101348 @default.
- W3035625410 hasConcept C138885662 @default.
- W3035625410 hasConcept C154945302 @default.
- W3035625410 hasConcept C162324750 @default.
- W3035625410 hasConcept C176217482 @default.
- W3035625410 hasConcept C177264268 @default.
- W3035625410 hasConcept C177769412 @default.
- W3035625410 hasConcept C199360897 @default.
- W3035625410 hasConcept C21547014 @default.
- W3035625410 hasConcept C2776401178 @default.
- W3035625410 hasConcept C2776760102 @default.
- W3035625410 hasConcept C41008148 @default.
- W3035625410 hasConcept C41895202 @default.
- W3035625410 hasConcept C43126263 @default.
- W3035625410 hasConcept C527412718 @default.
- W3035625410 hasConceptScore W3035625410C107673813 @default.
- W3035625410 hasConceptScore W3035625410C111919701 @default.
- W3035625410 hasConceptScore W3035625410C119857082 @default.
- W3035625410 hasConceptScore W3035625410C122770356 @default.
- W3035625410 hasConceptScore W3035625410C124101348 @default.
- W3035625410 hasConceptScore W3035625410C138885662 @default.
- W3035625410 hasConceptScore W3035625410C154945302 @default.
- W3035625410 hasConceptScore W3035625410C162324750 @default.
- W3035625410 hasConceptScore W3035625410C176217482 @default.
- W3035625410 hasConceptScore W3035625410C177264268 @default.
- W3035625410 hasConceptScore W3035625410C177769412 @default.
- W3035625410 hasConceptScore W3035625410C199360897 @default.
- W3035625410 hasConceptScore W3035625410C21547014 @default.
- W3035625410 hasConceptScore W3035625410C2776401178 @default.
- W3035625410 hasConceptScore W3035625410C2776760102 @default.
- W3035625410 hasConceptScore W3035625410C41008148 @default.
- W3035625410 hasConceptScore W3035625410C41895202 @default.
- W3035625410 hasConceptScore W3035625410C43126263 @default.
- W3035625410 hasConceptScore W3035625410C527412718 @default.
- W3035625410 hasLocation W30356254101 @default.
- W3035625410 hasLocation W30356254102 @default.
- W3035625410 hasLocation W30356254103 @default.
- W3035625410 hasOpenAccess W3035625410 @default.
- W3035625410 hasPrimaryLocation W30356254101 @default.
- W3035625410 hasRelatedWork W2049386694 @default.
- W3035625410 hasRelatedWork W2281360585 @default.
- W3035625410 hasRelatedWork W2380602769 @default.
- W3035625410 hasRelatedWork W2737894786 @default.
- W3035625410 hasRelatedWork W2774919641 @default.
- W3035625410 hasRelatedWork W2996333182 @default.
- W3035625410 hasRelatedWork W3011636446 @default.
- W3035625410 hasRelatedWork W3033962221 @default.
- W3035625410 hasRelatedWork W4210337481 @default.
- W3035625410 hasRelatedWork W4213149867 @default.
- W3035625410 isParatext "false" @default.
- W3035625410 isRetracted "false" @default.
- W3035625410 magId "3035625410" @default.
- W3035625410 workType "article" @default.