Mean Field Theory in Deep Metric Learning
This work addresses efficiency issues in deep metric learning for image retrieval, though it appears incremental as it builds on existing pair-based loss functions.
The paper tackles the high training complexity of conventional metric learning loss functions by applying mean field theory from statistical physics to deep metric learning, resulting in new loss functions that outperform baseline methods in two out of three image-retrieval datasets.
In this paper, we explore the application of mean field theory, a technique from statistical physics, to deep metric learning and address the high training complexity commonly associated with conventional metric learning loss functions. By adapting mean field theory for deep metric learning, we develop an approach to design classification-based loss functions from pair-based ones, which can be considered complementary to the proxy-based approach. Applying the mean field theory to two pair-based loss functions, we derive two new loss functions, MeanFieldContrastive and MeanFieldClassWiseMultiSimilarity losses, with reduced training complexity. We extensively evaluate these derived loss functions on three image-retrieval datasets and demonstrate that our loss functions outperform baseline methods in two out of the three datasets.