A Structured Prediction Approach for Label Ranking
This work addresses label ranking for machine learning applications, presenting an incremental improvement with specific embeddings to ease computational bottlenecks.
The paper tackles the label ranking problem by framing it as structured output regression, using a least squares surrogate loss and tailored feature embeddings to simplify the pre-image step, and demonstrates efficiency on real-world datasets.
We propose to solve a label ranking problem as a structured output regression task. We adopt a least square surrogate loss approach that solves a supervised learning problem in two steps: the regression step in a well-chosen feature space and the pre-image step. We use specific feature maps/embeddings for ranking data, which convert any ranking/permutation into a vector representation. These embeddings are all well-tailored for our approach, either by resulting in consistent estimators, or by solving trivially the pre-image problem which is often the bottleneck in structured prediction. We also propose their natural extension to the case of partial rankings and prove their efficiency on real-world datasets.