RECORD
SimPSI: A Simple Strategy to Preserve Spectral Information in Time Series Data Augmentation
Machine Learning · Submitted 10 Dec 2023
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Our experimental results support that SimPSI considerably enhances the performance of time series data augmentations by preserving core spectral information.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Data augmentation is a crucial component in training neural networks to overcome the limitation imposed by data size, and several techniques have been studied for time series. Although these techniques are effective in certain tasks, they have yet to be generalized to time series benchmarks. We find that current data augmentation techniques ruin the core information contained within the frequency domain. To address this issue, we propose a simple strategy to preserve spectral information (SimPSI) in time series data augmentation. SimPSI preserves the spectral information by mixing the original and augmented input spectrum weighted by a preservation map, which indicates the importance score of each frequency. Specifically, our experimental contributions are to build three distinct preservation maps: magnitude spectrum, saliency map, and spectrum-preservative map. We apply SimPSI to various time series data augmentations and evaluate its effectiveness across a wide range of time series benchmarks. Our experimental results support that SimPSI considerably enhances the performance of time series data augmentations by preserving core spectral information. The source code used in the paper is available at https://github.com/Hyun-Ryu/simpsi.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Data augmentation is a crucial component in training neural networks to overcome the limitation imposed by data size, and several techniques have been studied for time series. | background 1.00 | 0.05 | 0.02 | asserted 1.00 |
| s1 | Although these techniques are effective in certain tasks, they have yet to be generalized to time series benchmarks. | gap 0.97 | 0.07 | 0.03 | asserted 0.96 |
| s2 | We find that current data augmentation techniques ruin the core information contained within the frequency domain. | gap 0.52 | 0.89 | 0.12 | asserted 1.00 |
| s3 | To address this issue, we propose a simple strategy to preserve spectral information (SimPSI) in time series data augmentation. | contribution 0.95 | 0.78 | 0.08 | asserted 0.98 |
| s4 | SimPSI preserves the spectral information by mixing the original and augmented input spectrum weighted by a preservation map, which indicates the importance score of each frequency. | method 0.99 | 0.80 | 0.06 | asserted 1.00 |
| s5 | Specifically, our experimental contributions are to build three distinct preservation maps: magnitude spectrum, saliency map, and spectrum-preservative map. | contribution 0.80 | 0.94 | 0.03 | asserted 0.98 |
| s6 | We apply SimPSI to various time series data augmentations and evaluate its effectiveness across a wide range of time series benchmarks. | method 0.58 | 0.96 | 0.04 | asserted 0.64 |
| s7 | Our experimental results support that SimPSI considerably enhances the performance of time series data augmentations by preserving core spectral information. | result 0.91 | 0.96 | 0.90 | asserted 0.98 |
| s8 | The source code used in the paper is available at https://github.com/Hyun-Ryu/simpsi. | contribution 0.68 | 0.40 | 0.01 | asserted 0.97 |
What was examined, and what was not
- arXiv listing abstract v2
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Data augmentation is a crucial component in training neural networks to overcome the limitation imposed by data size, and several techniques have been studied for time series. Although these techniques are effective in certain tasks, they have yet to be generalized to time series benchmarks. We find that current data augmentation techniques ruin the core information contained within the frequency domain. To address this issue, we propose a simple strategy to preserve spectral information (SimPSI) in time series data augmentation. SimPSI preserves the spectral information by mixing the original and augmented input spectrum weighted by a preservation map, which indicates the importance score of each frequency. Specifically, our experimental contributions are to build three distinct preservation maps: magnitude spectrum, saliency map, and spectrum-preservative map. We apply SimPSI to various time series data augmentations and evaluate its effectiveness across a wide range of time series benchmarks. Our experimental results support that SimPSI considerably enhances the performance of time series data augmentations by preserving core spectral information. The source code used in the paper is available at https://github.com/Hyun-Ryu/simpsi.
Human review
How the literature uses it
1 arXiv paper cite this in 1 sentence
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.