Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387435293> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W4387435293 endingPage "312" @default.
- W4387435293 startingPage "301" @default.
- W4387435293 abstract "The prevalence of machine learning in biomedical research is rapidly growing, yet the trustworthiness of such research is often overlooked. While some previous works have investigated the ability of adversarial attacks to degrade model performance in medical imaging, the ability to falsely improve performance via recently-developed “enhancement attacks” may be a greater threat to biomedical machine learning. In the spirit of developing attacks to better understand trustworthiness, we developed two techniques to drastically enhance prediction performance of classifiers with minimal changes to features: 1) general enhancement of prediction performance, and 2) enhancement of a particular method over another. Our enhancement framework falsely improved classifiers’ accuracy from 50% to almost 100% while maintaining high feature similarities between original and enhanced data (Pearson’s $$r's>0.99$$ ). Similarly, the method-specific enhancement framework was effective in falsely improving the performance of one method over another. For example, a simple neural network outperformed logistic regression by 17% on our enhanced dataset, although no performance differences were present in the original dataset. Crucially, the original and enhanced data were still similar ( $$r=0.99$$ ). Our results demonstrate the feasibility of minor data manipulations to achieve any desired prediction performance, which presents an interesting ethical challenge for the future of biomedical machine learning. These findings emphasize the need for more robust data provenance tracking and other precautionary measures to ensure the integrity of biomedical machine learning research. The code is available at https://github.com/mattrosenblatt7/enhancement_EPIMI ." @default.
- W4387435293 created "2023-10-09" @default.
- W4387435293 creator A5032271246 @default.
- W4387435293 creator A5060428905 @default.
- W4387435293 creator A5069233580 @default.
- W4387435293 date "2023-01-01" @default.
- W4387435293 modified "2023-10-09" @default.
- W4387435293 title "Gradient-Based Enhancement Attacks in Biomedical Machine Learning" @default.
- W4387435293 cites W2011301426 @default.
- W4387435293 cites W2030561475 @default.
- W4387435293 cites W2069088601 @default.
- W4387435293 cites W2085561705 @default.
- W4387435293 cites W2104695005 @default.
- W4387435293 cites W2418392267 @default.
- W4387435293 cites W2526511911 @default.
- W4387435293 cites W2924551358 @default.
- W4387435293 cites W2952766316 @default.
- W4387435293 cites W2964043980 @default.
- W4387435293 cites W3021182036 @default.
- W4387435293 cites W3035751626 @default.
- W4387435293 cites W3150635270 @default.
- W4387435293 cites W3173556047 @default.
- W4387435293 cites W3203801990 @default.
- W4387435293 cites W3205084690 @default.
- W4387435293 cites W4247200422 @default.
- W4387435293 cites W4286488964 @default.
- W4387435293 cites W4312280786 @default.
- W4387435293 cites W4312662259 @default.
- W4387435293 cites W4322736917 @default.
- W4387435293 cites W4376602474 @default.
- W4387435293 cites W9657784 @default.
- W4387435293 doi "https://doi.org/10.1007/978-3-031-45249-9_29" @default.
- W4387435293 hasPublicationYear "2023" @default.
- W4387435293 type Work @default.
- W4387435293 citedByCount "0" @default.
- W4387435293 crossrefType "book-chapter" @default.
- W4387435293 hasAuthorship W4387435293A5032271246 @default.
- W4387435293 hasAuthorship W4387435293A5060428905 @default.
- W4387435293 hasAuthorship W4387435293A5069233580 @default.
- W4387435293 hasConcept C108583219 @default.
- W4387435293 hasConcept C119857082 @default.
- W4387435293 hasConcept C153701036 @default.
- W4387435293 hasConcept C154945302 @default.
- W4387435293 hasConcept C177264268 @default.
- W4387435293 hasConcept C199360897 @default.
- W4387435293 hasConcept C2776760102 @default.
- W4387435293 hasConcept C2778403875 @default.
- W4387435293 hasConcept C2986800882 @default.
- W4387435293 hasConcept C38652104 @default.
- W4387435293 hasConcept C41008148 @default.
- W4387435293 hasConcept C50644808 @default.
- W4387435293 hasConcept C71924100 @default.
- W4387435293 hasConcept C99508421 @default.
- W4387435293 hasConceptScore W4387435293C108583219 @default.
- W4387435293 hasConceptScore W4387435293C119857082 @default.
- W4387435293 hasConceptScore W4387435293C153701036 @default.
- W4387435293 hasConceptScore W4387435293C154945302 @default.
- W4387435293 hasConceptScore W4387435293C177264268 @default.
- W4387435293 hasConceptScore W4387435293C199360897 @default.
- W4387435293 hasConceptScore W4387435293C2776760102 @default.
- W4387435293 hasConceptScore W4387435293C2778403875 @default.
- W4387435293 hasConceptScore W4387435293C2986800882 @default.
- W4387435293 hasConceptScore W4387435293C38652104 @default.
- W4387435293 hasConceptScore W4387435293C41008148 @default.
- W4387435293 hasConceptScore W4387435293C50644808 @default.
- W4387435293 hasConceptScore W4387435293C71924100 @default.
- W4387435293 hasConceptScore W4387435293C99508421 @default.
- W4387435293 hasLocation W43874352931 @default.
- W4387435293 hasOpenAccess W4387435293 @default.
- W4387435293 hasPrimaryLocation W43874352931 @default.
- W4387435293 hasRelatedWork W1898221464 @default.
- W4387435293 hasRelatedWork W2374509987 @default.
- W4387435293 hasRelatedWork W3005856188 @default.
- W4387435293 hasRelatedWork W3163373470 @default.
- W4387435293 hasRelatedWork W3199750033 @default.
- W4387435293 hasRelatedWork W4235520445 @default.
- W4387435293 hasRelatedWork W4245118307 @default.
- W4387435293 hasRelatedWork W4327743613 @default.
- W4387435293 hasRelatedWork W4375867731 @default.
- W4387435293 hasRelatedWork W90316445 @default.
- W4387435293 isParatext "false" @default.
- W4387435293 isRetracted "false" @default.
- W4387435293 workType "book-chapter" @default.