Towards trustable SHAP scores
SHAP scores represent the proposed use of the well-known Shapley values in eXplainable Artificial Intelligence (XAI). Recent work has shown that the exact computation of SHAP scores can produce unsatisfactory results. Concretely, for some ML models, SHAP scores will mislead with respect to relative feature influence. To address these limitations, recently proposed alternatives exploit different axiomatic aggregations, all of which are defined in terms of abductive explanations. However, the proposed axiomatic aggregations are not Shapley values. This paper investigates how SHAP scores can be modified so as to extend axiomatic aggregations to the case of Shapley values in XAI. More importantly, the proposed new definition of SHAP scores avoids all the known cases where unsatisfactory results have been identified. The paper also characterizes the complexity of computing the novel definition of SHAP scores, highlighting families of classifiers for which computing these scores is tractable. Furthermore, the paper proposes modifications to the existing implementations of SHAP scores. These modifications eliminate some of the known limitations of SHAP scores, and have negligible impact in terms of performance.
Code (0)
등록된 구현이 없습니다.
Tasks
Explainable artificial intelligenceExplainable Artificial Intelligence (XAI)Methods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
A Unified Analytical Framework for Trustable Machine Learning and Automation Running with Blockchain
Traditional machine learning algorithms use data from databases that are mutable, and therefore the data cannot be fully trusted. Also, the machine learning process is difficult to automate. This paper proposes building …
BIG-bench Machine LearningA Privacy-Preserving and Trustable Multi-agent Learning Framework
Distributed multi-agent learning enables agents to cooperatively train a model without requiring to share their datasets. While this setting ensures some level of privacy, it has been shown that, even when data is not di…
Privacy PreservingReference Trustable Decoding: A Training-Free Augmentation Paradigm for Large Language Models
Large language models (LLMs) have rapidly advanced and demonstrated impressive capabilities. In-Context Learning (ICL) and Parameter-Efficient Fine-Tuning (PEFT) are currently two mainstream methods for augmenting LLMs t…
Few-Shot LearningIn-Context Learningparameter-efficient fine-tuningQuestion Answering+2ARNOR: Attention Regularization based Noise Reduction for Distant Supervision Relation Classification
Distant supervision is widely used in relation classification in order to create large-scale training data by aligning a knowledge base with an unlabeled corpus. However, it also introduces amounts of noisy labels where …
ClassificationGeneral ClassificationRelationRelation Classification+1Resilience of the Electric Grid through Trustable IoT-Coordinated Assets (Extended version)
The electricity grid has evolved from a physical system to a cyber-physical system with digital devices that perform measurement, control, communication, computation, and actuation. The increased penetration of distribut…