Deep Ordinal Regression using Optimal Transport Loss and Unimodal Output Probabilities

11/15/2020
by   Uri Shaham, et al.
0

We propose a framework for deep ordinal regression, based on unimodal output distribution and optimal transport loss. Despite being seemingly appropriate, in many recent works the unimodality requirement is either absent, or implemented using soft targets, which do not guarantee unimodal outputs at inference. In addition, we argue that the standard maximum likelihood objective is not suitable for ordinal regression problems, and that optimal transport is better suited for this task, as it naturally captures the order of the classes. Inspired by the well-known Proportional Odds model, we propose to modify its design by using an architectural mechanism which guarantees that the model output distribution will be unimodal. We empirically analyze the different components of our propose approach and demonstrate their contribution to the performance of the model. Experimental results on three real-world datasets demonstrate that our proposed approach performs on par with several recently proposed deep learning approaches for deep ordinal regression with unimodal output probabilities, while having guarantee on the output unimodality. In addition, we demonstrate that the level of prediction uncertainty of the model correlates with its accuracy.

READ FULL TEXT
research
10/14/2021

Order Constraints in Optimal Transport

Optimal transport is a framework for comparing measures whereby a cost i...
research
02/14/2022

Continuously Generalized Ordinal Regression for Linear and Deep Models

Ordinal regression is a classification task where classes have an order ...
research
09/14/2018

Entropic optimal transport is maximum-likelihood deconvolution

We give a statistical interpretation of entropic optimal transport by sh...
research
05/15/2019

Geometric Losses for Distributional Learning

Building upon recent advances in entropy-regularized optimal transport, ...
research
03/25/2021

Learning Probabilistic Ordinal Embeddings for Uncertainty-Aware Regression

Uncertainty is the only certainty there is. Modeling data uncertainty is...
research
05/10/2022

THOR: Threshold-Based Ranking Loss for Ordinal Regression

In this work, we present a regression-based ordinal regression algorithm...
research
03/03/2020

Regression via Implicit Models and Optimal Transport Cost Minimization

This paper addresses the classic problem of regression, which involves t...

Please sign up or login with your details

Forgot password? Click here to reset