Title | ||
---|---|---|
Direct Incorporation of L_1 -Regularization into Generalized Matrix Learning Vector Quantization. |
Abstract | ||
---|---|---|
Frequently, high-dimensional features are used to represent data to be classified. This paper proposes a new approach to learn interpretable classification models from such high-dimensional data representation. To this end, we extend a popular prototype-based classification algorithm, the matrix learning vector quantization, to incorporate an enhanced feature selection objective via (L_1)-regularization. In contrast to previous work, we propose a framework that directly optimizes this objective using the alternating direction method of multipliers (ADMM) and manifold optimization. We evaluate our method on synthetic data and on real data for speech-based emotion recognition. Particularly, we show that our method achieves state-of-the-art results on the Berlin Database of Emotional speech and show its abilities to select relevant dimensions from the eGeMAPS set of audio features. |
Year | Venue | Field |
---|---|---|
2018 | ICAISC | External Data Representation,Feature selection,Pattern recognition,Matrix (mathematics),Emotion recognition,Computer science,Learning vector quantization,Synthetic data,Regularization (mathematics),Artificial intelligence,Manifold |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
24 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Falko Lischke | 1 | 0 | 0.34 |
Thomas Neumann | 2 | 0 | 0.34 |
Sven Hellbach | 3 | 63 | 9.77 |
Thomas Villmann | 4 | 1279 | 118.19 |
Hans-Joachim Böhme | 5 | 143 | 20.86 |