Matches in Nanopublications for { ?s ?p ?o <https://w3id.org/np/RAAnumXKqMyA6FRjwlI0AWDgFT8rnYBqvgCwOHWrrhxT4/assertion>. }
Showing items 1 to 13 of
13
with 100 items per page.
- 10365540 type article assertion.
- 10365540 type FAIRDigitalObject assertion.
- 10365540 label "Dual Accuracy-Quality-Driven Neural Network for Prediction Interval Generation" assertion.
- 10365540 comment "Accurate uncertainty quantification is necessary to enhance the reliability of deep learning (DL) models in real-world applications. In the case of regression tasks, prediction intervals (PIs) should be provided along with the deterministic predictions of DL models. Such PIs are useful or “high-quality (HQ)” as long as they are sufficiently narrow and capture most of the probability density. In this article, we present a method to learn PIs for regression-based neural networks (NNs) automatically in addition to the conventional target predictions. In particular, we train two companion NNs: one that uses one output, the target estimate, and another that uses two outputs, the upper and lower bounds of the corresponding PI. Our main contribution is the design of a novel loss function for the PI-generation network that takes into account the output of the target-estimation network and has two optimization objectives: minimizing the mean PI width and ensuring the PI integrity using constraints that maximize the PI probability coverage implicitly. Furthermore, we introduce a self-adaptive coefficient that balances both objectives within the loss function, which alleviates the task of fine-tuning. Experiments using a synthetic dataset, eight benchmark datasets, and a real-world crop yield prediction dataset showed that our method was able to maintain a nominal probability coverage and produce significantly narrower PIs without detriment to its target estimation accuracy when compared to those PIs generated by three state-of-the-art neural-network-based methods. In other words, our method was shown to produce higher quality PIs. Major findings:The DualAQD framework produces "prediction intervals" that inform users of the confidence level associated with an AI model's specific estimate. This method generates narrower and more accurate confidence ranges than existing methods while maintaining high overall target accuracy. The system successfully identifies regions of high uncertainty in crop yield predictions, increasing the reliability of deep learning models for high-stakes decision-making." assertion.
- 10365540 creator 0000-0001-9487-5622 assertion.
- 10365540 creator 0000-0003-2911-8558 assertion.
- 10365540 subject topic_3316 assertion.
- 10365540 publisher 01n002310 assertion.
- 10365540 startDate "2022" assertion.
- 10365540 endDate "2023" assertion.
- 10365540 hasMetadata RAAnumXKqMyA6FRjwlI0AWDgFT8rnYBqvgCwOHWrrhxT4 assertion.
- 10365540 contactPoint "john.sheppard@montana.edu" assertion.
- 10365540 funder 02w0trx84 assertion.