Multi-target prediction: a unifying view on problems and methods

DOI10.1007/S10618-018-0595-5zbMATH Open1464.62399arXiv1809.02352OpenAlexW2891165828MaRDI QIDQ2218329FDOQ2218329

Authors: Willem Waegeman, Krzysztof Dembczyński, Eyke Hüllermeier 

Publication date: 15 January 2021

Published in: Data Mining and Knowledge Discovery (Search for Journal in Brave)

Abstract: Multi-target prediction (MTP) is concerned with the simultaneous prediction of multiple target variables of diverse type. Due to its enormous application potential, it has developed into an active and rapidly expanding research field that combines several subfields of machine learning, including multivariate regression, multi-label classification, multi-task learning, dyadic prediction, zero-shot learning, network inference, and matrix completion. In this paper, we present a unifying view on MTP problems and methods. First, we formally discuss commonalities and differences between existing MTP problems. To this end, we introduce a general framework that covers the above subfields as special cases. As a second contribution, we provide a structured overview of MTP methods. This is accomplished by identifying a number of key properties, which distinguish such methods and determine their suitability for different types of problems. Finally, we also discuss a few challenges for future research.

Full work available at URL: https://arxiv.org/abs/1809.02352

Recommendations

• Multi-target regression via input space expansion: treating targets as inputs
• Multi-target prediction for dummies using two-branch neural networks
• scientific article; zbMATH DE number 6276187
• scientific article; zbMATH DE number 5957307
• Multivariate prediction
• A Unified Approach to Some Prediction Problems

zbMATH Keywords

multivariate regressionmulti-task learningcollaborative filteringmulti-label classificationpairwise learningzero-shot learningdyadic predictionmulti-target prediction (MTP)

Mathematics Subject Classification ID

General nonlinear regression (62J02) Learning and adaptive systems in artificial intelligence (68T05) Hypothesis testing in multivariate analysis (62H15) Inference from stochastic processes and prediction (62M20)

Cites Work

• Spectral regularization algorithms for learning large incomplete matrices
• The elements of statistical learning. Data mining, inference, and prediction
• Title not available (Why is that?)
• Multi-output learning via spectral filtering
• Kernels for vector-valued functions: a review
• Reduced-rank regression for the multivariate linear model
• The ubiquitous Kronecker product
• Exact matrix completion via convex optimization
• Estimation with quadratic loss.
• Multi-dimensional classification with Bayesian networks
• Title not available (Why is that?)
• 10.1162/153244304322765658
• Low-rank matrix completion using alternating minimization
• Combining instance-based learning and logistic regression for multilabel classification
• Label ranking by learning pairwise preferences
• Multi-task learning for classification with Dirichlet process priors
• Learning intransitive reciprocal relations with kernel methods
• Decision trees for hierarchical multi-label classification
• A survey of hierarchical classification across different application domains
• A new approach to collaborative filtering: operator estimation with spectral regularization
• On label dependence and loss minimization in multi-label classification
• Title not available (Why is that?)
• Kernel-based learning of hierarchical multilabel classification models
• Learning multiple tasks with kernel methods
• Title not available (Why is that?)
• Multivariate regression analysis and canonical variates
• Multilabel classification with principal label space transformation
• Title not available (Why is that?)
• Multi-target regression via input space expansion: treating targets as inputs
• Dyad ranking using Plackett-Luce models based on joint feature representations
• Efficient regularized least-squares algorithms for conditional ranking on relational data
• On the Bayes-optimality of F-measure maximizers
• Machine Learning: ECML 2004

Cited In (11)

• Distributed algorithms of solving linear matrix equations via double-layered networks
• ML-PDA: advances and a new multitarget approach
• Exact and efficient top-\(K\) inference for multi-target prediction by querying separable linear relational models
• Title not available (Why is that?)
• Classifier chains: a review and perspectives
• Generalized vec trick for fast learning of pairwise kernel models
• Multi-target prediction for dummies using two-branch neural networks
• Title not available (Why is that?)
• Learning-augmented heuristics for scheduling parallel serial-batch processing machines
• Distributed solver for linear matrix inequalities: an optimization perspective
• Multi-target regression via input space expansion: treating targets as inputs

Uses Software

• NMF
• ElemStatLearn
• softImpute
• word2vec
• DeCAF
• OverFeat
• HCP

This page was built for publication: Multi-target prediction: a unifying view on problems and methods

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2218329)