Why Engine

Explainable AI in Manufacturing Industry

Why Engine is a new, experimental solution component that will allow the assistant to answer “why” questions concerning advices and predictions provided by the DIA. Explainability will increase trust in the applied AI-based functions.

The Why Engine will generate explanations of the provided predictions and recommendations to workers on the shop floor and to new workers in training. This is an exciting new application of explainability in the manufacturing context.

The Why Engine will communicate with the train data quality analytics models of the Prescriptive Quality Analytics component and interact with the Cognitive Advisor service. The basic rules and recommendations generated by the Cognitive Advisor will be used for its interpretation.

Video

How to acess and use the Why Engine Prototype

Public Deliverables

Why Engine Theory

Why Engine Prototype

Scientific Publications

N. Asher, J. Hunter, ‘When learning becomes impossible’, FACCT 2022, ACM Conference on Fairness, Accountability, and Transparency

A. Ignativ, Y. Izza, P.J. Stuckey & J. Marques-Silva: Using MaxSAT for Efficient Explanations of Tree Ensembles: Association for the Advancement of Artificial Intelligence (AAAI 2022)

Y. Izza, J. Marques-Silva, 2021, “On Explaining Random Forests with SAT”, International Joint Conference on Artificial Intelligence (IJCAI).

N. Asher, L. De Lara, S. Paul, C. RusselL.: (2022) Counterfactual Models for Fair and Adequate Explanations. Machine Learning and Knowledge Extraction

X. Huang, Y. Izza, A. Ignatiev, J. Marques-Silva : On Efficiently Explaining Graph-Based Classifiers (2021) KR 2021

P. Besse, E. del Barrio, P. Gordaliza, J.M. Loubes, L. Risser, : A survey of Bias in Machine Learning through Theorism of Statistical Parity. The American Statistician

N. Asher, J. Hunter & S. Paul. : Bias in Semantic and Discourse Interpretation (2021) Linguistics and Philosophy

G. Cabodi, P.E. Camurati, A. Ignatiev, J. Marques-Silva, M. Palena, & P. Pasini,: Optimizing binary decision diagrams for interpretable machine learning classification (2021) Design, Automation & Test in Europe Conference & Exhibition

N. Asher, S. Paul. C. Russell, ‘Fair and Adequate Explanations’, Workshop on Explainable AI, International Cross-Domain Conference for Machine Learning and Knowledge Extraction, 2021

A. Ignatiev and J. Marques-Silva: SAT-Based Rigorous Explanations for Decision Lists (2021). International Conference on Theory and Applications of Satisfiability Testing (SAT)

Y. Izza and J. Marques-Silva: On Explaining Random Forests with SAT (2021). International Joint Conference on Artificial Intelligence (IJCAI)

J. Marques-Silva, T. Gerspacher, M. C. Cooper, A. Ignatiev and N. Narodytska: Explanations for Monotonic Classifiers (2021). International Conference on Machine Learning (ICML)

A. Ignatiev, E.Lam, J. Marques-Silva & P. J. Stuckey: Reasoning-Based Learning of Interpretable ML Models (2021) International Joint Conference on Artificial Intelligence (IJCAI)

M.C. Cooper,: On the tractability of explaining decisions of classifiers.27th International Conference on Principles and Practice of Constraint Programming (CP 2021)

A.A. Shrotri, N. Narodytska, A. Ignatiev, J. Marques-Silva, K.S. Meel & M. Vardi : Constraint-Driven Explanations of Black-Box ML Models (2020) ICLR

T.L. Gouic, J.M. Loubes, P. Rigollet, : Projection to Fairness in Statistical Learning (2020)

GDPR Cookie Consent with Real Cookie Banner