Conceitos Básicos
The author proposes X-Selector to dynamically select explanations for AI predictions, aiming to guide users to better decisions based on the impact of different explanation combinations.
Resumo
The paper discusses the importance of selecting explanations in Intelligent Decision Support Systems (IDSSs) empowered by Artificial Intelligence (AI). It introduces X-Selector as a method to predict and select explanations that influence user decisions positively. The study includes experiments comparing X-Selector's performance with other strategies in a stock trading simulation, highlighting its potential benefits and challenges based on AI accuracy levels.
Estatísticas
IDSSs have shown promise in improving user decisions through XAI-generated explanations along with AI predictions.
The results suggest the potential of X-Selector to guide users to recommended decisions and improve performance when AI accuracy is high.
In the second experiment, we compared the results of explanations selected by X-Selector with ARGMAX and ALL.
The accuracy of StockAI for high-accuracy was 0.750, which was the highest, and that for low-accuracy was 0.333, the chance level of three-class classification.
Citações
"The results suggest that ARGMAX strategy works better with high AI accuracy, and ALL is more effective when AI accuracy is lower."
"X-Selector aims to take a further step to predict concrete decisions taking the effects of explanations into account and proactively influences them to improve the performance of each decision-making."