Add to ORKGInternational audienceThe goal of the work presented in this paper is to provide techniques to challenge the results of an algorithmic decision system relying on machine learning. We highlight the differences between explanations and justifications and outline a framework to generate evidence to support or to dismiss challenges. We also present the results of a preliminary study to assess users' perception of the different types of challenges proposed here and their benefits to detect incorrect results.Cet article présente des techniques pour contester les résultats d'un système algorithmique de décision basé sur l'apprentissage. Nous soulignons les différences entre les explications et les justifications et proposons un cadre pour générer ...
International audienceThe lack of validation and of explainability of some Machine Learning (ML) mod...
International audienceThe lack of explainability of machine learning (ML) techniques poses operation...
Recent work on interpretability in machine learning and AI has focused on the building of simplified...
Dans un contexte favorable à la rationalisation des décisions par des objectifs mesurables et des mé...
In a context favorable to the rationalization of decisions through measurable objectives and quantit...
The opacity of some recent Machine Learning (ML) techniques have raised fundamental questions on the...
In recent years, the use of algorithms coming from the Machine Learning literature has skyrocketed. ...
The thesis defended in this document starts with considering some formal approaches to Machine Perce...
The justification of an algorithm's outcomes is important in many domains, and in particular in the ...
The operations of deep networks are widely acknowledged to be inscrutable. The growing field of “Exp...
The field of machine learning has flourished over the past couple of decades. With huge amounts of d...
PhD ThesesThere is plenty of evidence that humans disagree on the interpretation of many tasks in N...
In AI and law, systems that are designed for decision support should be explainable when pursuing ju...
Black box AI systems for automated decision making, often based on machine learning over (big) data,...
International audienceThe lack of validation and of explainability of some Machine Learning (ML) mod...
International audienceThe lack of explainability of machine learning (ML) techniques poses operation...
Recent work on interpretability in machine learning and AI has focused on the building of simplified...
Dans un contexte favorable à la rationalisation des décisions par des objectifs mesurables et des mé...
In a context favorable to the rationalization of decisions through measurable objectives and quantit...
The opacity of some recent Machine Learning (ML) techniques have raised fundamental questions on the...
In recent years, the use of algorithms coming from the Machine Learning literature has skyrocketed. ...
The thesis defended in this document starts with considering some formal approaches to Machine Perce...
The justification of an algorithm's outcomes is important in many domains, and in particular in the ...
The operations of deep networks are widely acknowledged to be inscrutable. The growing field of “Exp...
The field of machine learning has flourished over the past couple of decades. With huge amounts of d...
PhD ThesesThere is plenty of evidence that humans disagree on the interpretation of many tasks in N...
In AI and law, systems that are designed for decision support should be explainable when pursuing ju...
Black box AI systems for automated decision making, often based on machine learning over (big) data,...
International audienceThe lack of validation and of explainability of some Machine Learning (ML) mod...
International audienceThe lack of explainability of machine learning (ML) techniques poses operation...
Recent work on interpretability in machine learning and AI has focused on the building of simplified...