We argue that, when establishing and benchmarking machine learning (ml)models, the research community should favour evaluation metrics that bettercapture the value delivered by their model in practical applications.
This paper proposes a novel visualisation approach to compare generated substantial amounts of machine learning (ml) models trained with a different number of features of a given data set while revealing implicit dependent relations such as feature importance for ml explanations.
The dependence of ml models with dynamic number of features is encoded into the structure of visualisation, where the dependence of ml models and their dependent features are directly revealed from related line connections.
Development of new machine learning models is typically done on manuallycurated data sets, making them unsuitable for evaluating the models'performance during operations, where the evaluation needs to be performed automatically on incoming streams of new data.
With this in mind, we developed a web-based visualization system that allows the users to quickly gather headline performance numbers while maintaining confidence that the underlying data pipeline is functioning properly.
This paper formally models the strategic repeated interactions between a machine learning (ml) model and associated explanationmethod, and an end-user who is seeking a prediction/label and its explanationfor a query/input, by means of game theory.
In this game, a malicious end-user must strategically decide when to stop querying and attempt to compromise the system, while the system must strategically decide how much information (in the form of noisy explanations) it should share with the end-user and when to stopsharing, all without knowing the type (honest/malicious) of the end-user.
This paper conducts an in-depth literaturereview of a large volume of research papers that focused on the qualityassurance of machine learning (ml) models into software systems.
We developed a taxonomy of quality assurance issues of machine learning software applications (mlsas) by mapping the various ml adoption challenges across different phases of software development life cycles (sdlc).
Magnetism prediction is of great significance for metallic glasses, which have shown great commercial value.
In this work, machine learning (ml) models learned from a large amount of experimental data were trained based on extreme gradient boosting (xgboost), artificial neural networks (ann), and random forest to predict the magnetic properties of metallic glasses.