{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/multi-scale-convolutional-neural-networks-for","title":"Multi-Scale Convolutional Neural Networks for Time Series Classification","arxiv_id":"1603.06995","date":"2016-03-22","proceeding":null,"authors":["Zhicheng Cui","Wenlin Chen","Yixin Chen"],"abstract":"Time series classification (TSC), the problem of predicting class labels of\ntime series, has been around for decades within the community of data mining\nand machine learning, and found many important applications such as biomedical\nengineering and clinical prediction. However, it still remains challenging and\nfalls short of classification accuracy and efficiency. Traditional approaches\ntypically involve extracting discriminative features from the original time\nseries using dynamic time warping (DTW) or shapelet transformation, based on\nwhich an off-the-shelf classifier can be applied. These methods are ad-hoc and\nseparate the feature extraction part with the classification part, which limits\ntheir accuracy performance. Plus, most existing methods fail to take into\naccount the fact that time series often have features at different time scales.\nTo address these problems, we propose a novel end-to-end neural network model,\nMulti-Scale Convolutional Neural Networks (MCNN), which incorporates feature\nextraction and classification in a single framework. Leveraging a novel\nmulti-branch layer and learnable convolutional layers, MCNN automatically\nextracts features at different scales and frequencies, leading to superior\nfeature representation. MCNN is also computationally efficient, as it naturally\nleverages GPU computing. We conduct comprehensive empirical evaluation with\nvarious existing methods on a large number of benchmark datasets, and show that\nMCNN advances the state-of-the-art by achieving superior accuracy performance\nthan other leading methods.","url_abs":"http://arxiv.org/abs/1603.06995v4","url_pdf":"http://arxiv.org/pdf/1603.06995v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"multi-scale-convolutional-neural-networks-for","repo_url":"https://github.com/aidatalab314/AUO_forecasting","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"multi-scale-convolutional-neural-networks-for","repo_url":"https://github.com/zdcuob/Fully-Convlutional-Neural-Networks-for-state-of-the-art-time-series-classification-","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"dynamic-time-warping","task_name":"Dynamic Time Warping"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series","task_name":"Time Series Analysis"},{"task_slug":"time-series-classification","task_name":"Time Series Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1603.06995","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}