Improving Deep Representation Learning Via Auxiliary Learnable Target Coding

Abstract

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 sev

Stats

• citations5
• S2 citations—
• github stars0
• HF likes0
• heat score5.84
• arxiv keyliu2023improving

Related papers

• Unsupervised Representation Learning Via Neural Activation Coding (2021)1.56
• Deep Class-wise Hashing: Semantics-preserving Hashing Via Class-wise Loss (2018)11.19
• Adaptive Additive Classification-based Loss For Deep Metric Learning (2020)0.00
• Anti-collapse Loss For Deep Metric Learning Based On Coding Rate Metric (2024)0.00
• Dual Asymmetric Deep Hashing Learning (2018)9.03
• A Quadruplet Loss For Enforcing Semantically Coherent Embeddings In Multi-output Classification Problems (2020)6.77
• Compact Hash Code Learning With Binary Deep Neural Network (2017)9.03
• Fast Training Of Triplet-based Deep Binary Embedding Networks (2016)14.62