A survey on multi‐output regression

Abstract

In recent years, a plethora of approaches have been proposed to deal with the increasingly challenging task of multi-output regression. This paper provides a survey on state-of-the-art multi-output regression methods, that are categorized as problem transformation and algorithm adaptation methods. In addition, we present the mostly used performance evaluation measures, publicly available data sets for multi-output regression real-world problems, as well as open-source software frameworks.

A novel multi-target regression framework for time-series prediction of drug efficacy

“…who used multi-output support vector regression to simultaneously estimate the different biophysical parameters from remote sensing images, and it is also used to predict the wind noise intensity of vehicle components23. Hanen et al 24. categorized state-of-the-art multi-target regression methods as transformation methods and algorithm adaptation methods.…”

A locally weighted learning method based on a data gravitation model for multi-target regression

“…, (x n , y n )} of n training examples, the goal in multi-target regression problems is to learn a predictive model that, given an unseen input vector x, is able to predict a target vector y that best approximates the true target vector y. 4,38 Up to date, a large number of methods have been proposed to resolve multi-target regression problems. The taxonomy of multi-target regression algorithms can be organised into two groups: problem transformation methods and algorithm adaptation methods 4 .…”

“…4,38 Up to date, a large number of methods have been proposed to resolve multi-target regression problems. The taxonomy of multi-target regression algorithms can be organised into two groups: problem transformation methods and algorithm adaptation methods 4 . Problem transformation methods transform a multi-target regression problem into several single-target regression problems.…”

“…Multi-target regression concerns the task of predicting multiple continuous variables using a common set of input variables 4,38 . Multi-target regressors are commonly characterized by learning a global model to fit all of the training data.…”

Expressive Control of Singing Voice Synthesis Using Musical Contexts and a Parametric F0 Model

“…Thus, a multi-output decision tree [19] is built for each segment's type. All contexts listed above for the F0 are used for building the trees, except for "attacks" and "releases" segments for which only the pitch and length of the current note (respectively first and last note of a musical phrase) are considered.…”