Treat it as a statistical model

**Have you checked the [list of proposed rules](https://github.com/Benjamin-Lee/deep-rules/issues?q=is%3Aissue+is%3Aopen+label%3Arule) to see if the rule has already been proposed?**

- [x] Yes

This rule is related to @ttriche's comments in #5. Because DL models can be difficult to interpret intuitively, there is a temptation to anthropomorphize DL models. We should resist this temptation. We would not expect a linear model to learn beyond pattern recognition, so we should also not expect an overparameterized non-linear model to do so. Because of this, we need to pay just as much attention to statistical caveats with DL models as we do with traditional models. For instance:

- Don't use a deep learning model to extrapolate beyond the domain of the training set. While architectures with saturating nonlinearities will produce outputs in a reasonable range outside the training domain, that does not mean the predictions are reliable.
- Don't use predictive accuracy or interpretability as evidence of casual reasoning (#9). We wouldn't interpret R^2 or MSE as evidence of causal reasoning in statistical models; don't do it for DL models either.
- Interrogate the model for input sensitivity. Does your model respond to confounding effects, translations on the data domain, etc.? Do you want it to? Just as we would inspect the coefficients of a linear model, we should inspect the sensitivity of DL models to understand which signals they identify.

Start by reading the comments in #5 and the proposed-rules list linked in the issue, then compare this proposal with any existing rules. Done means the project has an agreed rule addressing statistical caveats in deep-learning models, including extrapolation, causal claims, and input sensitivity.

Written by the indexing model from the issue text.