# Compositionality in Time Series: A Proof of Concept using Symbolic Dynamics and Compositional Data Augmentation

arXiv: https://arxiv.org/abs/2508.20656 · HTML: https://mlsanity.com/paper/2508.20656/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 55 references that are in the map.
- First posted on arXiv 28 Aug 2025.

## Builds on (most cited first, 25 of 25)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations](https://mlsanity.com/paper/1811.12359/) (2018)
- [Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs](https://mlsanity.com/paper/1706.02633/) (2017)
- [Measuring Compositional Generalization: A Comprehensive Method on Realistic Data](https://mlsanity.com/paper/1912.09713/) (2019)
- [COGS: A Compositional Generalization Challenge Based on Semantic Interpretation](https://mlsanity.com/paper/2010.05465/) (2020)
- [Time Series Data Augmentation for Deep Learning: A Survey](https://mlsanity.com/paper/2002.12478/) (2020)
- [Good-Enough Compositional Data Augmentation](https://mlsanity.com/paper/1904.09545/) (2019)
- [Set Functions for Time Series](https://mlsanity.com/paper/1909.12064/) (2019)
- [Self-Supervised Transformer for Sparse and Irregularly Sampled Multivariate Clinical Time-Series](https://mlsanity.com/paper/2107.14293/) (2021)
- [Learning to Recombine and Resample Data for Compositional Generalization](https://mlsanity.com/paper/2010.03706/) (2020)
- [Compositional Generalization from First Principles](https://mlsanity.com/paper/2307.05596/) (2023)
- [Visual Representation Learning Does Not Generalize Strongly Within the Same Domain](https://mlsanity.com/paper/2107.08221/) (2021)
- [Improving Compositional Generalization with Latent Structure and Data Augmentation](https://mlsanity.com/paper/2112.07610/) (2021)
- [Compositional Generalization in Unsupervised Compositional Representation Learning: A Study on Disentanglement and Emergent Language](https://mlsanity.com/paper/2210.00482/) (2022)
- [Transformer Grammars: Augmenting Transformer Language Models with Syntactic Inductive Biases at Scale](https://mlsanity.com/paper/2203.00633/) (2022)
- [A Survey of Mix-based Data Augmentation: Taxonomy, Methods, Applications, and Explainability](https://mlsanity.com/paper/2212.10888/) (2022)
- [Towards Generating Real-World Time Series Data](https://mlsanity.com/paper/2111.08386/) (2021)
- [Revisiting the Compositional Generalization Abilities of Neural Sequence Models](https://mlsanity.com/paper/2203.07402/) (2022)
- [Motiflets -- Simple and Accurate Detection of Motifs in Time Series](https://mlsanity.com/paper/2206.03735/) (2022)
- [A Self-Supervised Learning-based Approach to Clustering Multivariate Time-Series Data with Missing Values (SLAC-Time): An Application to TBI Phenotyping](https://mlsanity.com/paper/2302.13457/) (2023)
- [Robust Augmentation for Multivariate Time Series Classification](https://mlsanity.com/paper/2201.11739/) (2022)
- [Empirical Study of Mix-based Data Augmentation Methods in Physiological Time Series Data](https://mlsanity.com/paper/2309.09970/) (2023)
- [Early Prediction of Causes (not Effects) in Healthcare by Long-Term Clinical Time Series Forecasting](https://mlsanity.com/paper/2408.03816/) (2024)
- [Towards Equipping Transformer with the Ability of Systematic Compositionality](https://mlsanity.com/paper/2312.07280/) (2023)

## Abstract (arXiv)

This work investigates whether time series of natural phenomena can be understood as being generated by sequences of latent states which are ordered in systematic and regular ways. We focus on clinical time series and ask whether clinical measurements can be interpreted as being generated by meaningful physiological states whose succession follows systematic principles. Uncovering the underlying compositional structure will allow us to create synthetic data to alleviate the notorious problem of sparse and low-resource data settings in clinical time series forecasting, and deepen our understanding of clinical data. We start by conceptualizing compositionality for time series as a property of the data generation process, and then study data-driven procedures that can reconstruct the elementary states and composition rules of this process. We evaluate the success of this methods using two empirical tests originating from a domain adaptation perspective. Both tests infer the similarity of the original time series distribution and the synthetic time series distribution from the similarity of expected risk of time series forecasting models trained and tested on original and synthesized data in specific ways. Our experimental results show that the test set performance achieved by training on compositionally synthesized data is comparable to training on original clinical time series data, and that evaluation of models on compositionally synthesized test data shows similar results to evaluating on original test data, outperforming randomization-based data augmentation. An additional downstream evaluation of the prediction task of sequential organ failure assessment (SOFA) scores shows significant performance gains when model training is entirely based on compositionally synthesized data compared to training on original data.