Summary of Cost-sensitive Learning to Defer to Multiple Experts with Workload Constraints, by Jean V. Alves et al.
Cost-Sensitive Learning to Defer to Multiple Experts with Workload Constraints
by Jean V. Alves, Diogo Leitão, Sérgio Jesus, Marco O. P. Sampaio, Javier Liébana, Pedro Saleiro, Mário A. T. Figueiredo, Pedro Bizarro
First submitted to arxiv on: 11 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed DeCCaF framework addresses key limitations in existing learning to defer (L2D) approaches by incorporating cost-sensitive scenarios, reduced data requirements, and human work-capacity constraints. The authors employ supervised learning to model the probability of human error and constraint programming to minimize the error cost while considering workload limitations. Experimental results demonstrate a significant reduction in misclassification cost, achieving an average 8.4% improvement compared to baselines. This approach has potential applications in fraud detection and other domains where decision-making involves trade-offs between accuracy and computational resources. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Learning to defer is a new way for humans and artificial intelligence (AI) to work together more effectively. Right now, AI systems often make decisions that are not as good as what a human would do. The problem is that current approaches to learning to defer have some major limitations. For example, they don’t account for situations where one type of mistake is worse than another, or where the AI has to make many predictions in a short amount of time. To address these issues, researchers developed a new approach called DeCCaF. This framework uses machine learning and computer programming to help the AI decide when it should make a decision and when it’s better to ask a human for help. The results show that this approach can significantly reduce the number of mistakes made by the AI. |
Keywords
* Artificial intelligence * Machine learning * Probability * Supervised
