# Quantile Regression using Neural Networks (Custom Loss function)

I would like to perform a Quantile Regression within Mathematica's neural network framework.

Since there exist a couple of tutorials for this exact problem in Tensorflow, i.e. Deep Quantile Regression, I know this can be done by defining the right loss function for this particular problem, i.e.

$$\rho_\tau(u)= \max[ u\tau, u(\tau-1)].$$

Then a specific quantile $$\tau$$ can be found by minimizing the expected loss of $$y-f(X)$$ with respect to $$f(X)$$, where $$f(X)$$ is the predicted (quantile) model and $$y$$ is the observed value for the corresponding input $$X$$.

My question then, is it possible to define a custom loss function for NetTrain[], and if so, how can this be done?

An example of the data can be loaded by,

data = Flatten[Uncompress[Import["https://pastebin.com/raw/G2kdCu4i"]]]


Note that in this particular setting the observed values are 3-tuples.

Disclaimer, this is my first time using Mathematica's neural network framework.

• Yes, you can, check the documentation for LossFunction. It can be a little limited however. It might be helpful if you give some example of the output data and the target data. Oct 12 '18 at 12:41
• @CarlLange I did came across the documentation for LossFunction but it only used built-in layers which threw me off. I also added an example of my data. Oct 12 '18 at 13:14
• It looks like your QuantileLossLayer answers your question - you could post it as an answer and close the question! Oct 12 '18 at 18:08

Going through the documentation of LossFunction it dawned on me that I needed to define a custom Layer via NetGraph, hence

QuantileLossLayer[τ_] := NetGraph[<|

net = NetChain[{8, Tanh, 16, Tanh, 3}];

• What is the net, net? Oct 17 '18 at 12:24
• @AntonAntonov net is any neural network you deem fit, but I get your point and added an working example. Oct 17 '18 at 18:05