Multi-Prototype Label Ranking with Novel Pairwise-to-Total-Rank Aggregation

Grbovic, Mihajlo (Yahoo! Labs, USA) | Djuric, Nemanja (Temple University, USA) | Vucetic, Slobodan (Temple University, USA)

AAAI Conferences 

We propose a multi-prototype-based algorithm for online learning of soft pairwise-preferences over labels. The algorithm learns soft label preferences via minimization of the proposed soft rank-loss measure, and can learn from total orders as well as from various types of partial orders. The soft pairwise preference algorithm outputs are further aggregated to produce a total label ranking prediction using a novel aggregation algorithm that outperforms existing aggregation solutions. Experiments on synthetic and real-world data demonstrate state-of-the-art performance of the proposed model.