all AI news
A unifying partially-interpretable framework for neural network-based extreme quantile regression. (arXiv:2208.07581v2 [stat.ML] UPDATED)
stat.ML updates on arXiv.org arxiv.org
Risk management in many environmental settings requires an understanding of
the mechanisms that drive extreme events. Useful metrics for quantifying such
risk are extreme quantiles of response variables conditioned on predictor
variables that describe e.g., climate, biosphere and environmental states.
Typically these quantiles lie outside the range of observable data and so, for
estimation, require specification of parametric extreme value models within a
regression framework. Classical approaches in this context utilise linear or
additive relationships between predictor and response variables …
arxiv framework ml network neural network quantile regression