Review ML Target

A target is the desired output, label, behavior, or outcome a model is trying to predict or reach.

In machine learning, a target is the value or outcome a model is trained to predict. It may be a class label, number, sequence, action, or another desired output.

Choosing a target turns a real-world goal into a prediction problem. If the target is only a proxy, the model may learn to optimize the proxy while missing the outcome people care about.

A maintenance model uses “component failed within 30 days” as its target rather than the easier but less useful target “technician opened a ticket.”

The target is not necessarily the true goal. It is the measurable outcome supplied to the learning process and may be incomplete, delayed, or noisy.

The target is what the model is asked to predict.

* [[output|Output]] * [[training-data|Training Data]] * [[loss-function|Loss Function]] * [[evaluation-metric|Evaluation Metric]]

* [Google for Developers, “Machine Learning Glossary”](https://developers.google.com/machine-learning/glossary) — provides established boundaries for core machine-learning vocabulary.

* Compare [[output|Output]], [[training-data|Training Data]], [[loss-function|Loss Function]], and [[evaluation-metric|Evaluation Metric]] to place ML Target in the surrounding system. * Follow the cited source material above for the method, standard, or research context behind this draft.