Some Theoretical Limitations of t-SNE
MIT
arXiv:2604.1329510.5h-index: 61
AI Analysis
For practitioners using t-SNE for visualization, this work formalizes known limitations but is incremental in nature.
This paper establishes mathematical results showing how t-SNE can lose important data features, providing a theoretical framework for understanding its limitations.
t-SNE has gained popularity as a dimension reduction technique, especially for visualizing data. It is well-known that all dimension reduction techniques may lose important features of the data. We provide a mathematical framework for understanding this loss for t-SNE by establishing a number of results in different scenarios showing how important features of data are lost by using t-SNE.