Learning Interpretable Models Using Uncertainty Oracles · HackerPapers discussions

Learning Interpretable Models Using Uncertainty Oracles — mentioned by abhgh

Sat, 30 Nov 2024 08:44:12 GMT

Yes, that's the one: https://arxiv.org/pdf/1906.06852

Paper and discussion history. A mention is not an endorsement.

Learning Interpretable Models Using Uncertainty Oracles — mentioned by aspenmayer

Sat, 30 Nov 2024 08:34:41 GMT

Is a preprint of your paper available? I looked at your blog a bit and was able to find this, which may be it? > Learning Interpretable Models Using Uncertainty Oracles https://arxiv.org/abs/1906.06852 https://doi.org/10.48550/arXiv.1906.06852

Paper and discussion history. A mention is not an endorsement.

Learning Interpretable Models Using Uncertainty Oracles — mentioned by abhgh

Sun, 07 Apr 2024 05:46:31 GMT

Not for DL models, but I was exploring doing this in a specific setting (I've posted about this earlier). For small-sized models (for some reasonable definition of size, e.g., depth of a decision tree, or # trees in a gradient boosting forest, or # non-zero coefficients in a linear model), I realized that you could make them even smaller while retaining their accuracy by selectively presenting training data to them, i.e., ignore some training data points, repeat certain others. See [1] - the x-axis shows the original model size and the y-axis shows the model size obtained by this process at…

Paper and discussion history. A mention is not an endorsement.