English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this paper we address the application of pre-processing techniques tomulti-channel time series data with varying lengths, which we refer to as thealignment problem, for downstream machine learning. The misalignment ofmulti-channel time series data may occur for a variety of reasons, such asmissing data, varying sampling rates, or inconsistent collection times. Weconsider multi-channel time series data collected from the MIT SuperCloud HighPerformance Computing (HPC) center, where different job start times and varyingrun times of HPC jobs result in misaligned data. This misalignment makes itchallenging to build AI/ML approaches for tasks such as compute workloadclassification. Building on previous supervised classification work with theMIT SuperCloud Dataset, we address the alignment problem via three broad, lowoverhead approaches: sampling a fixed subset from a full time series,performing summary statistics on a full time series, and sampling a subset ofcoefficients from time series mapped to the frequency domain. Our bestperforming models achieve a classification accuracy greater than 95%,outperforming previous approaches to multi-channel time series classificationwith the MIT SuperCloud Dataset by 5%. These results indicate our low overheadapproaches to solving the alignment problem, in conjunction with standardmachine learning techniques, are able to achieve high levels of classificationaccuracy, and serve as a baseline for future approaches to addressing thealignment problem, such as kernel methods.

An Evaluation of Low Overhead Time Series Preprocessing Techniques for Downstream Machine Learning