ROCKET: exceptionally fast and accurate time series classification using random convolutional kernels

@article{Dempster2020ROCKETEF,
  title={ROCKET: exceptionally fast and accurate time series classification using random convolutional kernels},
  author={Angus Dempster and Franccois Petitjean and Geoffrey I. Webb},
  journal={Data Mining and Knowledge Discovery},
  year={2020},
  volume={34},
  pages={1454-1495}
}
Most methods for time series classification that attain state-of-the-art accuracy have high computational complexity, requiring significant training time even for smaller datasets, and are intractable for larger datasets. Additionally, many existing methods focus on a single type of feature such as shape or frequency. Building on the recent success of convolutional neural networks for time series classification, we show that simple linear classifiers using random convolutional kernels achieve… 

MiniRocket: A Very Fast (Almost) Deterministic Transform for Time Series Classification

TLDR
This work reformulates Rocket into a new method, MiniRocket, which is up to 75 times faster than Rocket on larger datasets, and almost deterministic (and optionally, fully deterministic), while maintaining essentially the same accuracy.

MultiRocket: Effective summary statistics for convolutional outputs in time series classification

TLDR
It is shown that it is possible to significantly improve the accuracy of MiniRocket (and Rocket), with some additional computational expense, by expanding the set of features produced by the transform, making MultiRocket (for MiniRocket with Multiple Features) overall the single most accurate method on the datasets in the UCR archive.

The Temporal Dictionary Ensemble (TDE) Classifier for Time Series Classification

TLDR
A further extension of these dictionary based classifiers that combines the best elements of the others combined with a novel approach to constructing ensemble members based on an adaptive Gaussian process model of the parameter space is proposed.

An Examination of the State-of-the-Art for Multivariate Time Series Classification

TLDR
This work investigates the state-of-the-art for multivariate time series classification using the UEA MTSC benchmark, and proposes a simple statistics-based time series classifier that is as accurate as DTW, but is an order of magnitude faster.

The great multivariate time series classification bake off: a review and experimental evaluation of recent algorithmic advances

TLDR
It is demonstrated that four classifiers are significantly more accurate than the benchmark dynamic time warping algorithm and that one of these recently proposed classifiers achieves significant improvement on the archive datasets in at least an order of magnitude less time than the other three.

A Transformer-based Framework for Multivariate Time Series Representation Learning

TLDR
A novel framework for multivariate time series representation learning based on the transformer encoder architecture, which can offer substantial performance benefits over fully supervised learning on downstream tasks, both with but even without leveraging additional unlabeled data, i.e., by reusing the existing data samples.

Time series classification for predictive maintenance on event logs

TLDR
This paper introduces a new data set for predictive maintenance on automated teller machines (ATMs) log data and compares the performance of different representation methods for predicting the occurrence of a breakdown, comparing the state-of-the-art representation and classification methods.

Monash University, UEA, UCR Time Series Regression Archive

TLDR
The first TSR benchmarking archive is introduced, which contains 19 datasets from different domains, with varying number of dimensions, unequal length dimensions, and missing values, and an initial benchmark on existing models is done.

Autoencoder-based Representation Learning from Heterogeneous Multivariate Time Series Data of Mechatronic Systems

TLDR
This paper presents a method for unsupervised feature extraction using autoencoder networks that specifically addresses the heterogeneous nature of the database and reduces the amount of labeled training data required compared to existing methods.

HIVE-COTE 2.0: a new meta ensemble for time series classification

TLDR
This work proposes comprehensive changes to the HIVE-COTE algorithm which significantly improve its accuracy and usability, and introduces two novel classifiers, the Temporal Dictionary Ensemble and Diverse Representation Canonical Interval Forest, which replace existing ensemble members.
...

References

SHOWING 1-10 OF 71 REFERENCES

Multi-Scale Convolutional Neural Networks for Time Series Classification

TLDR
A novel end-to-end neural network model, Multi-Scale Convolutional Neural Networks (MCNN), which incorporates feature extraction and classification in a single framework, leading to superior feature representation.

InceptionTime: Finding AlexNet for Time Series Classification

TLDR
An important step towards finding the AlexNet network for TSC is taken by presenting InceptionTime---an ensemble of deep Convolutional Neural Network models, inspired by the Inception-v4 architecture, which outperforms HIVE-COTE's accuracy together with scalability.

Deep learning for time series classification: a review

TLDR
This article proposes the most exhaustive study of DNNs for TSC by training 8730 deep learning models on 97 time series datasets and provides an open source deep learning framework to the TSC community.

Deep Neural Network Ensembles for Time Series Classification

TLDR
This paper shows how an ensemble of 60 deep learning models can significantly improve upon the current state-of-the-art performance of neural networks for TSC, when evaluated over the UCR/UEA archive: the largest publicly available benchmark for time series analysis.

Interpretable time series classification using linear models and multi-resolution multi-domain symbolic representations

TLDR
A multi-resolution multi-domain linear classifier achieves a similar accuracy to the state-of-the-art COTE ensemble, and to recent deep learning methods (FCN, ResNet), but uses a fraction of the time and memory required by either COTE or deep models.

Time series classification from scratch with deep neural networks: A strong baseline

TLDR
The proposed Fully Convolutional Network (FCN) achieves premium performance to other state-of-the-art approaches and the exploration of the very deep neural networks with the ResNet structure is also competitive.

TS-CHIEF: a scalable and accurate forest algorithm for time series classification

TLDR
A novel TSC algorithm, TS-CHIEF (Time Series Combination of Heterogeneous and Integrated Embedding Forest), which rivals HIVE-COTE in accuracy but requires only a fraction of the runtime.

Binary Shapelet Transform for Multiclass Time Series Classification

TLDR
A one vs all encoding scheme is evaluated which simplifies the quality assessment calculations, speeds up the execution through facilitating more frequent early abandon and increases accuracy for multi-class problems.

The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances

TLDR
This work implemented 18 recently proposed algorithms in a common Java framework and compared them against two standard benchmark classifiers (and each other) by performing 100 resampling experiments on each of the 85 datasets, indicating that only nine of these algorithms are significantly more accurate than both benchmarks.

Scalable Dictionary Classifiers for Time Series Classification

TLDR
Changes to the way BOSS chooses classifiers for its ensemble are evaluated, replacing its parameter search with random selection, achieving a significant reduction in build time without a significant change in accuracy on average when compared to BOSS.
...