A Hierarchical Multi-Output Nearest Neighbor Model for Multi-Output Dependence Learning
This addresses a generalization of classification for scenarios with dependent outputs, though it appears incremental as it builds on nearest neighbor methods.
The paper tackles the problem of Multi-Output Dependence learning, where multiple dependent outputs exist for inputs, by introducing the Hierarchical Multi-Output Nearest Neighbor model, which uses basic models per output and a nearest neighbor refinement to address this relation approximation task.
Multi-Output Dependence (MOD) learning is a generalization of standard classification problems that allows for multiple outputs that are dependent on each other. A primary issue that arises in the context of MOD learning is that for any given input pattern there can be multiple correct output patterns. This changes the learning task from function approximation to relation approximation. Previous algorithms do not consider this problem, and thus cannot be readily applied to MOD problems. To perform MOD learning, we introduce the Hierarchical Multi-Output Nearest Neighbor model (HMONN) that employs a basic learning model for each output and a modified nearest neighbor approach to refine the initial results.