Usually, uncertainties of Machine Learning predictions are just regarded as a sign of poor prediction accuracy or as a consequence of lacking input features. This talk illustrates how modeling uncertainties can improve ML based decisions.
The uncertainty of predictions from ML models is often just regarded as a bad thing - as a sign of model deficiencies, too little data, or lacking input features. In this talk, I will argue why knowing this uncertainty can be valuable. This especially applies if your ML model is used to minimize some kind of cost - either implicitly by a human or explicitly by an optimization algorithm.
The first step in realizing this value is going beyond point estimators towards predicting a probability distribution. In this talk, I will review how this can be done with scikit-learn and related tools.
The second step is feeding these predictions into an optimization algorithm as probability distributions. As an example, I will present grocery store orders that are calculated from demand predictions. In extreme cases, including the uncertainty of the ML predictions turns out to be necessary for the optimization to be useful in practice at all.
Algorithms, Data Science, Machine Learning, Statistics
Domain Expertise:expert
Python Skill Level:basic
Abstract as a tweet:Usually, uncertainties of Machine Learning predictions are just regarded as a sign of poor prediction accuracy or as a consequence of lacking input features. This talk illustrates how modeling uncertainties can improve ML based decisions.