An Efficient Approach to Informative Feature Extraction from Multimodal Data
This work addresses a bottleneck in multimodal data analysis for researchers and practitioners by providing an incremental improvement over existing correlation-based methods.
The paper tackled the limitation of strict whitening constraints in HGR maximal correlation for multimodal feature extraction by proposing Soft-HGR, a framework that avoids these constraints while preserving feature geometry, resulting in more informative feature mappings and efficient optimization as shown empirically.
One primary focus in multimodal feature extraction is to find the representations of individual modalities that are maximally correlated. As a well-known measure of dependence, the Hirschfeld-Gebelein-Rényi (HGR) maximal correlation becomes an appealing objective because of its operational meaning and desirable properties. However, the strict whitening constraints formalized in the HGR maximal correlation limit its application. To address this problem, this paper proposes Soft-HGR, a novel framework to extract informative features from multiple data modalities. Specifically, our framework prevents the "hard" whitening constraints, while simultaneously preserving the same feature geometry as in the HGR maximal correlation. The objective of Soft-HGR is straightforward, only involving two inner products, which guarantees the efficiency and stability in optimization. We further generalize the framework to handle more than two modalities and missing modalities. When labels are partially available, we enhance the discriminative power of the feature representations by making a semi-supervised adaptation. Empirical evaluation implies that our approach learns more informative feature mappings and is more efficient to optimize.