The Earth Mover's Pinball Loss: Quantiles for Histogram-Valued Regression

by   Florian List, et al.

Although ubiquitous in the sciences, histogram data have not received much attention by the Deep Learning community. Whilst regression and classification tasks for scalar and vector data are routinely solved by neural networks, a principled approach for estimating histogram labels as a function of an input vector or image is lacking in the literature. We present a dedicated method for Deep Learning-based histogram regression, which incorporates cross-bin information and yields distributions over possible histograms, expressed by τ-quantiles of the cumulative histogram in each bin. The crux of our approach is a new loss function obtained by applying the pinball loss to the cumulative histogram, which for 1D histograms reduces to the Earth Mover's distance (EMD) in the special case of the median (τ = 0.5), and generalizes it to arbitrary quantiles. We validate our method with an illustrative toy example, a football-related task, and an astrophysical computer vision problem. We show that with our loss function, the accuracy of the predicted median histograms is very similar to the standard EMD case (and higher than for per-bin loss functions such as cross-entropy), while the predictions become much more informative at almost no additional computational cost.



There are no comments yet.


page 7

page 15


Smooth Loss Functions for Deep Top-k Classification

The top-k error is a common measure of performance in machine learning a...

Improved Training Speed, Accuracy, and Data Utilization Through Loss Function Optimization

As the complexity of neural network models has grown, it has become incr...

Sparse Coding with Earth Mover's Distance for Multi-Instance Histogram Representation

Sparse coding (Sc) has been studied very well as a powerful data represe...

Continuous Histogram Loss: Beyond Neural Similarity

Similarity learning has gained a lot of attention from researches in rec...

Niose-Sampling Cross Entropy Loss: Improving Disparity Regression Via Cost Volume Aware Regularizer

Recent end-to-end deep neural networks for disparity regression have ach...

DeepMoM: Robust Deep Learning With Median-of-Means

Data used in deep learning is notoriously problematic. For example, data...

Calibrating Deep Neural Networks using Focal Loss

Miscalibration – a mismatch between a model's confidence and its correct...

Code Repositories


The Earth Mover's Pinball Loss: Quantiles for Histogram-Valued Regression

view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.