A General Framework for Visualizing Embedding Spaces of\titlebreak Neural Survival Analysis Models Based on Angular Information

George H Chen
Proceedings of the Conference on Health, Inference, and Learning, PMLR 209:440-476, 2023.

Abstract

We propose a general framework for visualizing any intermediate embedding representation used by any neural survival analysis model. Our framework is based on so-called \emph{anchor directions} in an embedding space. We show how to estimate these anchor directions using clustering or, alternatively, using user-supplied “concepts” defined by collections of raw inputs (e.g., feature vectors all from female patients could encode the concept “female”). For tabular data, we present visualization strategies that reveal how anchor directions relate to raw clinical features and to survival time distributions. We then show how these visualization ideas extend to handling raw inputs that are images. Our framework is built on looking at angles between vectors in an embedding space, where there could be “information loss” by ignoring magnitude information. We show how this loss results in a “clumping” artifact that appears in our visualizations, and how to reduce this information loss in practice.

Cite this Paper

BibTeX
@InProceedings{pmlr-v209-chen23b, title = {A General Framework for Visualizing Embedding Spaces of\titlebreak Neural Survival Analysis Models Based on Angular Information}, author = {Chen, George H}, booktitle = {Proceedings of the Conference on Health, Inference, and Learning}, pages = {440--476}, year = {2023}, editor = {Mortazavi, Bobak J. and Sarker, Tasmie and Beam, Andrew and Ho, Joyce C.}, volume = {209}, series = {Proceedings of Machine Learning Research}, month = {22 Jun--24 Jun}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v209/chen23b/chen23b.pdf}, url = {https://proceedings.mlr.press/v209/chen23b.html}, abstract = {We propose a general framework for visualizing any intermediate embedding representation used by any neural survival analysis model. Our framework is based on so-called \emph{anchor directions} in an embedding space. We show how to estimate these anchor directions using clustering or, alternatively, using user-supplied “concepts” defined by collections of raw inputs (e.g., feature vectors all from female patients could encode the concept “female”). For tabular data, we present visualization strategies that reveal how anchor directions relate to raw clinical features and to survival time distributions. We then show how these visualization ideas extend to handling raw inputs that are images. Our framework is built on looking at angles between vectors in an embedding space, where there could be “information loss” by ignoring magnitude information. We show how this loss results in a “clumping” artifact that appears in our visualizations, and how to reduce this information loss in practice.} }
Endnote
%0 Conference Paper %T A General Framework for Visualizing Embedding Spaces of\titlebreak Neural Survival Analysis Models Based on Angular Information %A George H Chen %B Proceedings of the Conference on Health, Inference, and Learning %C Proceedings of Machine Learning Research %D 2023 %E Bobak J. Mortazavi %E Tasmie Sarker %E Andrew Beam %E Joyce C. Ho %F pmlr-v209-chen23b %I PMLR %P 440--476 %U https://proceedings.mlr.press/v209/chen23b.html %V 209 %X We propose a general framework for visualizing any intermediate embedding representation used by any neural survival analysis model. Our framework is based on so-called \emph{anchor directions} in an embedding space. We show how to estimate these anchor directions using clustering or, alternatively, using user-supplied “concepts” defined by collections of raw inputs (e.g., feature vectors all from female patients could encode the concept “female”). For tabular data, we present visualization strategies that reveal how anchor directions relate to raw clinical features and to survival time distributions. We then show how these visualization ideas extend to handling raw inputs that are images. Our framework is built on looking at angles between vectors in an embedding space, where there could be “information loss” by ignoring magnitude information. We show how this loss results in a “clumping” artifact that appears in our visualizations, and how to reduce this information loss in practice.
APA
Chen, G.H.. (2023). A General Framework for Visualizing Embedding Spaces of\titlebreak Neural Survival Analysis Models Based on Angular Information. Proceedings of the Conference on Health, Inference, and Learning, in Proceedings of Machine Learning Research 209:440-476 Available from https://proceedings.mlr.press/v209/chen23b.html.

Related Material