Greybox XAI: A Neural-Symbolic learning framework to produce interpretable predictions for image classification

Bennetot, Adrien; Franchi, Gianni; Ser, Javier Del; Chatila, Raja; Díaz-Rodríguez, Natalia

doi:10.1016/j.knosys.2022.109947

Cited by 16 publications

(15 citation statements)

References 78 publications

(90 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we introduce the experiments carried out to validate the proposed methods (see Section 3) on the datasets presented in Section 4.1. We kept the splits in training and test sets that were proposed in [3,6].…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…Other authors have proposed to train object detectors or segmentation models as concept extractors to localize object parts that are used as concepts. The final model solution combines those models with a classifier that bases its decision on the detected object parts [3,6].…”

Section: Related Workmentioning

confidence: 99%

“…flat arch, pointed arch, porthole...) were also generated. The classification and analysis of those key elements can be used to explain the decision of a classifier as done in [3,21].…”

Section: Datasetsmentioning

confidence: 99%

“…Furthermore, the images are part-annotated on more than 40 different elements (i.e Leg, Body, Wheel,...). This dataset has already been explored on concept-based or part-based research articles [3,6].…”

Section: Datasetsmentioning

confidence: 99%

“…In this section, we compare our models and results to five state of the art approaches that were introduced above in Section 2. Three of the models are transparent models (Greybox [3], EXPLANet [6]) and [28] and the other two models are opaque models (DeiT-B [3,33] and MonuNet [21]). MonuNet is an ad-hoc solution for monument-style classification, which is why results are not available for the PASCAL dataset.…”

Section: Use Case: User Interventionmentioning

confidence: 99%

See 4 more Smart Citations

Concept logic trees: enabling user interaction for transparent image classification and human-in-the-loop learning

Rodríguez,

Cuéllar,

Morales

2024

Appl Intell

View full text Add to dashboard Cite

Interpretable deep learning models are increasingly important in domains where transparent decision-making is required. In this field, the interaction of the user with the model can contribute to the interpretability of the model. In this research work, we present an innovative approach that combines soft decision trees, neural symbolic learning, and concept learning to create an image classification model that enhances interpretability and user interaction, control, and intervention. The key novelty of our method relies on the fusion of an interpretable architecture with neural symbolic learning, allowing the incorporation of expert knowledge and user interaction. Furthermore, our solution facilitates the inspection of the model through queries in the form of first-order logic predicates. Our main contribution is a human-in-the-loop model as a result of the fusion of neural symbolic learning and an interpretable architecture. We validate the effectiveness of our approach through comprehensive experimental results, demonstrating competitive performance on challenging datasets when compared to state-of-the-art solutions.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

“…flat arch, pointed arch, porthole...) were also generated. The classification and analysis of those key elements can be used to explain the decision of a classifier as done in [3,21].…”

Section: Datasetsmentioning

confidence: 99%

Section: Datasetsmentioning

confidence: 99%

Section: Use Case: User Interventionmentioning

confidence: 99%

See 3 more Smart Citations

Concept logic trees: enabling user interaction for transparent image classification and human-in-the-loop learning

Rodríguez,

Cuéllar,

Morales

2024

Appl Intell

View full text Add to dashboard Cite

show abstract

Explainable artificial intelligence in breast cancer detection and risk prediction: A systematic scoping review

Ghasemi,

Hashtarkhani,

Schwartz

et al. 2024

Cancer Innovation

View full text Add to dashboard Cite

With the advances in artificial intelligence (AI), data‐driven algorithms are becoming increasingly popular in the medical domain. However, due to the nonlinear and complex behavior of many of these algorithms, decision‐making by such algorithms is not trustworthy for clinicians and is considered a black‐box process. Hence, the scientific community has introduced explainable artificial intelligence (XAI) to remedy the problem. This systematic scoping review investigates the application of XAI in breast cancer detection and risk prediction. We conducted a comprehensive search on Scopus, IEEE Explore, PubMed, and Google Scholar (first 50 citations) using a systematic search strategy. The search spanned from January 2017 to July 2023, focusing on peer‐reviewed studies implementing XAI methods in breast cancer datasets. Thirty studies met our inclusion criteria and were included in the analysis. The results revealed that SHapley Additive exPlanations (SHAP) is the top model‐agnostic XAI technique in breast cancer research in terms of usage, explaining the model prediction results, diagnosis and classification of biomarkers, and prognosis and survival analysis. Additionally, the SHAP model primarily explained tree‐based ensemble machine learning models. The most common reason is that SHAP is model agnostic, which makes it both popular and useful for explaining any model prediction. Additionally, it is relatively easy to implement effectively and completely suits performant models, such as tree‐based models. Explainable AI improves the transparency, interpretability, fairness, and trustworthiness of AI‐enabled health systems and medical devices and, ultimately, the quality of care and outcomes.

show abstract

Learner Modeling Interpretability and Explainability in Intelligent Adaptive Systems

Zapata-Rivera,

Arslan

2024

Integrated Science

View full text Add to dashboard Cite

Greybox XAI: A Neural-Symbolic learning framework to produce interpretable predictions for image classification

Cited by 16 publications

References 78 publications

Concept logic trees: enabling user interaction for transparent image classification and human-in-the-loop learning

Concept logic trees: enabling user interaction for transparent image classification and human-in-the-loop learning

Explainable artificial intelligence in breast cancer detection and risk prediction: A systematic scoping review

Learner Modeling Interpretability and Explainability in Intelligent Adaptive Systems

Contact Info

Product

Resources

About