Improving Deep Representation Learning via Auxiliary Learnable Target Coding
This work addresses the limitation of inflexible target codes in semantic classification for machine learning practitioners, offering an incremental improvement over existing methods.
The paper tackles the problem of deep representation learning by introducing a learnable target coding method to replace pre-defined codes like one-hot, resulting in improved discriminative representations with larger between-class margins and better handling of imbalanced data, as demonstrated on visual classification and retrieval benchmarks.
Deep representation learning is a subfield of machine learning that focuses on learning meaningful and useful representations of data through deep neural networks. However, existing methods for semantic classification typically employ pre-defined target codes such as the one-hot and the Hadamard codes, which can either fail or be less flexible to model inter-class correlation. In light of this, this paper introduces a novel learnable target coding as an auxiliary regularization of deep representation learning, which can not only incorporate latent dependency across classes but also impose geometric properties of target codes into representation space. Specifically, a margin-based triplet loss and a correlation consistency loss on the proposed target codes are designed to encourage more discriminative representations owing to enlarging between-class margins in representation space and favoring equal semantic correlation of learnable target codes respectively. Experimental results on several popular visual classification and retrieval benchmarks can demonstrate the effectiveness of our method on improving representation learning, especially for imbalanced data. Source codes are made publicly available at \href{https://github.com/AkonLau/LTC}{https://github.com/AkonLau/LTC}.