Bad And Good Errors Valueweighted Skill Scores In Deep Ensemble Learning

# Bad and good errors: value-weighted skill scores in deep ensemble learning

**arXiv ID:** 2103.02881
**Authors:** Sabrina Guastavino, Michele Piana, Federico Benvenuto
**Published:** 2021-03-04T08:05:13Z
**Abstract:**
In this paper we propose a novel approach to realize forecast verification. Specifically, we introduce a strategy for assessing the severity of forecast errors based on the evidence that, on the one hand, a false alarm just anticipating an occurring event is better than one in the middle of consecutive non-occurring events, and that, on the other hand, a miss of an isolated event has a worse impact than a miss of a single event, which is part of several consecutive occurrences. Relying on this idea, we introduce a novel definition of confusion matrix and skill scores giving greater importance to the value of the prediction rather than to its quality. Then, we introduce a deep ensemble learning procedure for binary classification, in which the probabilistic outcomes of a neural network are clustered via optimization of these value-weighted skill scores. We finally show the performances of this approach in the case of three applications concerned with pollution, space weather and stock prize forecasting.

## Skill Description

This skill is generated from the arXiv paper: Bad and good errors: value-weighted skill scores in deep ensemble learning (2103.02881).

## How to Use

[To be filled in by the user or by future automation]

## References

- [arXiv:2103.02881](http://arxiv.org/abs/2103.02881v1)