What is it about?
Distance metrics are functions that provide a way to quantify how far apart two elements of a given set are to each other. So, we propose a new data representation space called Similarity space (S-space) that separates regions where similar/dissimilar objects lie together and help the convergence of the model.
Photo by Darling Arias on Unsplash
Why is it important?
In our daily lives, we can identify several patterns in the world around us. Human perception (along with our inferential ability) observes these patterns and helps us to recognize common features among collections of objects. Thus, identifying common characteristics among objects becomes a critical task for understanding the world and the things surrounding it.
Read the Original
This page is a summary of: A New Similarity Space Tailored for Supervised Deep Metric Learning, ACM Transactions on Intelligent Systems and Technology, November 2022, ACM (Association for Computing Machinery), DOI: 10.1145/3559766.
You can read the full text:
We propose a novel deep metric learning method. Differently from many works on this area, we defined a novel latent space obtained through an autoencoder. The new space, namely S-space, is divided into different regions that describe the positions where pairs of objects are similar/dissimilar. We locate makers to identify these regions. We estimate the similarities between objects through a kernel-based t-student distribution to measure the markers' distance and the new data representation. In our approach, we simultaneously estimate the markers' position in the S-space and represent the objects in the same space. Moreover, we propose a new regularization function to avoid similar markers to collapse altogether. We present evidences that our proposal can represent complex spaces, for instance, when groups of similar objects are located in disjoint regions. We compare our proposal to 9 different distance metric learning approaches (four of them are based on deep-learning) on 28 real-world heterogeneous datasets. According to the four quantitative metrics used, our method overcomes all the nine strategies from the literature.
The following have contributed to this page