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

Abstract

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 state-of-the-art accuracy with a fraction of the computational expense of existing methods.

27 pages, 23 figures

References in corpus (1)

• InceptionTime: Finding AlexNet for Time Series Classification

Cited by in corpus (50)

• Benchmarking Multivariate Time Series Classification Algorithms
• MINIROCKET: A Very Fast (Almost) Deterministic Transform for Time Series Classification
• HIVE-COTE 2.0: a new meta ensemble for time series classification
• Improving Position Encoding of Transformers for Multivariate Time Series Classification
• Bake off redux: a review and experimental evaluation of recent time series classification algorithms
• Deep Learning for Satellite Image Time Series Analysis: A Review
• Time-Series Classification in Smart Manufacturing Systems: An Experimental Evaluation of State-of-the-Art Machine Learning Algorithms
• The Temporal Dictionary Ensemble (TDE) Classifier for Time Series Classification
• A tale of two toolkits, report the third: on the usage and performance of HIVE-COTE v1.0
• ClaSP -- Parameter-free Time Series Segmentation
• LiteHAR: Lightweight Human Activity Recognition from WiFi Signals with Random Convolution Kernels
• The FreshPRINCE: A Simple Transformation Based Pipeline Time Series Classifier
• Dimensionality Reduction of Longitudinal 'Omics Data using Modern Tensor Factorization
• Random Dilated Shapelet Transform: A New Approach for Time Series Shapelets
• Privacy-Preserving Electricity Theft Detection based on Blockchain
• dCAM: Dimension-wise Class Activation Map for Explaining Multivariate Data Series Classification
• Time-Series Forecasting in Smart Manufacturing Systems: An Experimental Evaluation of the State-of-the-art Algorithms
• An ensemble of convolution-based methods for fault detection using vibration signals
• Unsupervised Feature Based Algorithms for Time Series Extrinsic Regression
• A Systematic Literature Review of Spatio-Temporal Graph Neural Network Models for Time Series Forecasting and Classification
• Unsupervised Multivariate Time-Series Transformers for Seizure Identification on EEG
• ChatTS: Aligning Time Series with LLMs via Synthetic Data for Enhanced Understanding and Reasoning
• Robust Explainer Recommendation for Time Series Classification
• SeqNAS: Neural Architecture Search for Event Sequence Classification
• AI-driven Java Performance Testing: Balancing Result Quality with Testing Time
• RandomNet: Clustering Time Series Using Untrained Deep Neural Networks
• Convolutional and Deep Learning based techniques for Time Series Ordinal Classification
• Automatic Feature Engineering for Time Series Classification: Evaluation and Discussion
• Estimating Motor Symptom Presence and Severity in Parkinson's Disease from Wrist Accelerometer Time Series using ROCKET and InceptionTime
• LIFE: Learning Individual Features for Multivariate Time Series Prediction with Missing Values
• Multivariate Functional Linear Discriminant Analysis for the Classification of Short Time Series with Missing Data
• COSCO: A Sharpness-Aware Training Framework for Few-shot Multivariate Time Series Classification
• Interpretable Feature Construction for Time Series Extrinsic Regression
• A Dictionary-based approach to Time Series Ordinal Classification
• Motor Imagery EEG Signal Classification Using Minimally Random Convolutional Kernel Transform and Hybrid Deep Learning
• ATCN: Resource-Efficient Processing of Time Series on Edge
• An Empirical Evaluation of Multivariate Time Series Classification with Input Transformation across Different Dimensions
• RED CoMETS: An ensemble classifier for symbolically represented multivariate time series
• Few Labels are all you need: A Weakly Supervised Framework for Appliance Localization in Smart-Meter Series
• A 1-D CNN inference engine for constrained platforms
• Unraveling Anomalies in Time: Unsupervised Discovery and Isolation of Anomalous Behavior in Bio-regenerative Life Support System Telemetry
• Adaptive Law-Based Transformation (ALT): A Lightweight Feature Representation for Time Series Classification
• Rock the KASBA: Blazingly Fast and Accurate Time Series Clustering
• Deep Learning for Time Series Classification of Parkinson's Disease Eye Tracking Data
• Binary Classification of Light and Dark Time Traces of a Transition Edge Sensor Using Convolutional Neural Networks
• Enhancing deep learning models for time series classification via knowledge distillation
• MIHT: A Hoeffding Tree for Time Series Classification using Multiple Instance Learning
• HIT-ROCKET: Hadamard-vector Inner-product Transformer for ROCKET
• ProtoTSNet: Interpretable Multivariate Time Series Classification With Prototypical Parts
• Mid infrared spectroscopy and milk quality traits: a data analysis competition at the "International Workshop on Spectroscopy and Chemometrics 2021"