{"id":{"repo_id":"mit","oai_identifier":"oai:dspace.mit.edu:1721.1/144671"},"canonical_url":"https://search.dev.ndltd.org/etd/mit/oai:dspace.mit.edu:1721.1/144671","repository":{"repo_id":"mit","name":"MIT","base_url":"https://dspace.mit.edu/oai/request"},"display":{"title":"Time Series Anomaly Detection using Prediction-Reconstruction Mixture Errors","abstract":"Anomaly detection on time series data is increasingly common across various industrial domains that require monitoring metrics to prevent potential accidents and economic losses. The complications of anomaly detection revolve around a scarcity of labeled data and the need to learn temporal correlations between multiple variables. Most successful unsupervised methods either use single-timestamp prediction or reconstruct entire time series. However, these methods are not mutually exclusive and can each offer complementary perspectives. This work first explores the successes and limitations of prediction-based and reconstruction-based methods. Next, it compares the effect of attention-based architectures with LSTM-based architectures on existing models. Finally, this research proposes a novel autoencoder architecture capable of producing bi-directional predictions while simultaneously reconstructing the original time series by optimizing a joint objective function. An ablation study using a mixture of prediction and reconstruction errors demonstrates that this simple architecture outperforms other state-of-the-art models for anomaly detection on both univariate and multivariate time series.","abstract_html":"Anomaly detection on time series data is increasingly common across various industrial domains that require monitoring metrics to prevent potential accidents and economic losses. The complications of anomaly detection revolve around a scarcity of labeled data and the need to learn temporal correlations between multiple variables. Most successful unsupervised methods either use single-timestamp prediction or reconstruct entire time series. However, these methods are not mutually exclusive and can each offer complementary perspectives. This work first explores the successes and limitations of prediction-based and reconstruction-based methods. Next, it compares the effect of attention-based architectures with LSTM-based architectures on existing models. Finally, this research proposes a novel autoencoder architecture capable of producing bi-directional predictions while simultaneously reconstructing the original time series by optimizing a joint objective function. An ablation study using a mixture of prediction and reconstruction errors demonstrates that this simple architecture outperforms other state-of-the-art models for anomaly detection on both univariate and multivariate time series.","abstract_has_math":false,"creators":["Wong, Lawrence C."],"institution":"Massachusetts Institute of Technology","degree_name":"Master","degree_level":null,"degree_discipline":null,"degree_department":"Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science","school":null,"contributors":[],"advisors":["Veeramachaneni, Kalyan"],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-05","date_published":"2022-05","updated_at":"2026-07-22T22:22:27Z","subjects":[],"languages":[],"rights":["In Copyright - Educational Use Permitted","Copyright MIT"],"rights_urls":["http://rightsstatements.org/page/InC-EDU/1.0/"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1721.1/144671","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Veeramachaneni, Kalyan"]},{"key":"dc:contributor.department","label":"Department","values":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"]},{"key":"dc:creator","label":"Author","values":["Wong, Lawrence C."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2022-08-29T16:03:41Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2022-08-29T16:03:41Z"]},{"key":"dc:date.issued","label":"Date","values":["2022-05"]},{"key":"dc:publisher","label":"Institution","values":["Massachusetts Institute of Technology"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master","Master of Engineering in Computer Science and Molecular Biology"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["In Copyright - Educational Use Permitted","Copyright MIT"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://rightsstatements.org/page/InC-EDU/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1721.1/144671"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Anomaly detection on time series data is increasingly common across various industrial domains that require monitoring metrics to prevent potential accidents and economic losses. The complications of anomaly detection revolve around a scarcity of labeled data and the need to learn temporal correlations between multiple variables. Most successful unsupervised methods either use single-timestamp prediction or reconstruct entire time series. However, these methods are not mutually exclusive and can each offer complementary perspectives. This work first explores the successes and limitations of prediction-based and reconstruction-based methods. Next, it compares the effect of attention-based architectures with LSTM-based architectures on existing models. Finally, this research proposes a novel autoencoder architecture capable of producing bi-directional predictions while simultaneously reconstructing the original time series by optimizing a joint objective function. An ablation study using a mixture of prediction and reconstruction errors demonstrates that this simple architecture outperforms other state-of-the-art models for anomaly detection on both univariate and multivariate time series."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["M.Eng."]},{"key":"dc:title","label":"Title","values":["Time Series Anomaly Detection using Prediction-Reconstruction Mixture Errors"]}]}],"canonical_facts":{"dc:contributor.advisor":["Veeramachaneni, Kalyan"],"dc:contributor.department":["Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science"],"dc:creator":["Wong, Lawrence C."],"dc:date.accessioned":["2022-08-29T16:03:41Z"],"dc:date.available":["2022-08-29T16:03:41Z"],"dc:date.issued":["2022-05"],"dc:description.abstract":["Anomaly detection on time series data is increasingly common across various industrial domains that require monitoring metrics to prevent potential accidents and economic losses. The complications of anomaly detection revolve around a scarcity of labeled data and the need to learn temporal correlations between multiple variables. Most successful unsupervised methods either use single-timestamp prediction or reconstruct entire time series. However, these methods are not mutually exclusive and can each offer complementary perspectives. This work first explores the successes and limitations of prediction-based and reconstruction-based methods. Next, it compares the effect of attention-based architectures with LSTM-based architectures on existing models. Finally, this research proposes a novel autoencoder architecture capable of producing bi-directional predictions while simultaneously reconstructing the original time series by optimizing a joint objective function. An ablation study using a mixture of prediction and reconstruction errors demonstrates that this simple architecture outperforms other state-of-the-art models for anomaly detection on both univariate and multivariate time series."],"dc:description.degree":["M.Eng."],"dc:identifier.uri":["https://hdl.handle.net/1721.1/144671"],"dc:publisher":["Massachusetts Institute of Technology"],"dc:rights":["In Copyright - Educational Use Permitted","Copyright MIT"],"dc:rights.uri":["http://rightsstatements.org/page/InC-EDU/1.0/"],"dc:title":["Time Series Anomaly Detection using Prediction-Reconstruction Mixture Errors"],"dc:type":["Thesis"],"thesis:degree_name":["Master","Master of Engineering in Computer Science and Molecular Biology"]},"updated_at":"2026-07-22T22:22:27Z"}