Waegeman, W., Dembczyński, K. & Hüllermeier, E. Multi-target prediction: a unifying view on problems and methods. Data Min Knowl Disc 33, 293–324 (2019). https://doi.org/10.1007/s10618-018-0595-5
Driven by FURP(FoSE Undergraduate Research Placement) Programme.
Publisher: Springer Nature
Expert Systems with Applications
Authors:
Willem Waegeman, Krzysztof Dembczyński, Eyke Hüllermeier
Index Terms - Extreme Learning Machine - Regularization - Multi-target regression - Robust to outliers - Alternating direction method of multipliers
DOI: 10.1007/s10618-018-0595-5
Background(Key Point):
Many problem settings in machine learning are concerned with the simultaneous prediction of multiple target variables of diverse type. Amongst others, such problem settings arise in multivariate regression, multi-label classification, multi-task learning, dyadic prediction, zero-shot learning, network inference, and matrix completion. These subfields of machine learning are typically studied in isolation, without highlighting or exploring important relationships.
Methodology:
Present a unifying view on what we call multi-target prediction (MTP) problems and methods.
Introduce a general framework that covers the above subfields as special cases, is accomplished by identifying a number of key properties, which distinguish such methods and determine their suitability for different types of problems.
Key Findings:
Note: This paper is not a typical review paper
Intend to focus on some general principles that might be helpful in identifying the right approach for a given problem;