{"id":{"repo_id":"ets-quebec","oai_identifier":"oai:espace.etsmtl.ca:2913"},"canonical_url":"https://search.ndltd.org/etd/ets-quebec/oai:espace.etsmtl.ca:2913","repository":{"repo_id":"ets-quebec","name":"ETS (Quebec)","base_url":"https://espace.etsmtl.ca/cgi/oai2"},"display":{"title":"Forecasting lightpath quality of transmission and implementing uncertainty in the forecast models","abstract":"The recent popularity of using Deep Learning models for the forecasting of time series calls for methods to not only predict the target but also to measure the uncertainty of the prediction accurately. Working with time series requires reliable forecasters. An intuitive N-Beats architecture was implemented in this thesis with the desirable properties of being fast to train and adaptable, without requiring modifications for the various time series array settings. Its performance was compared with other models, namely Long Short-Term Memory (LSTM) Multi-layer Perceptron (MLP) and a naïve method. The influence of different hyperparameters were investigated as to how the parameters affected the performances of the models. Our NBeats model outperformed well-known time series forecaster, LSTM. All the implementations were conducted in Python programming language. Random sampling was performed to avoid overfitting. Our target field data are Microsoft and NASP data sets. Estimating uncertainty is a component of reliable machine leaning models for time series forecasting. To measure the uncertainty and confidence level of the selected MLP model, Monte Carlo dropout, which approximates Bayesian uncertainty, was applied during inference. Quantile Regression was also implemented on the MLP algorithm as a baseline to predict the confidence intervals and to evaluate our strategy for estimating uncertainty. As a result, Bayesian approximation models indicated the necessity of calibration to adjust the predicted probability.","abstract_html":"The recent popularity of using Deep Learning models for the forecasting of time series calls for methods to not only predict the target but also to measure the uncertainty of the prediction accurately. Working with time series requires reliable forecasters. An intuitive N-Beats architecture was implemented in this thesis with the desirable properties of being fast to train and adaptable, without requiring modifications for the various time series array settings. Its performance was compared with other models, namely Long Short-Term Memory (LSTM) Multi-layer Perceptron (MLP) and a naïve method. The influence of different hyperparameters were investigated as to how the parameters affected the performances of the models. Our NBeats model outperformed well-known time series forecaster, LSTM. All the implementations were conducted in Python programming language. Random sampling was performed to avoid overfitting. Our target field data are Microsoft and NASP data sets. Estimating uncertainty is a component of reliable machine leaning models for time series forecasting. To measure the uncertainty and confidence level of the selected MLP model, Monte Carlo dropout, which approximates Bayesian uncertainty, was applied during inference. Quantile Regression was also implemented on the MLP algorithm as a baseline to predict the confidence intervals and to evaluate our strategy for estimating uncertainty. As a result, Bayesian approximation models indicated the necessity of calibration to adjust the predicted probability.","abstract_has_math":false,"creators":["Yousefi, Somaiieh"],"institution":"École de technologie supérieure","degree_name":"Mémoire de maîtrise électronique","degree_level":"masters","degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2021,"date_issued":"2021-12-13","date_published":"2021-12-13","updated_at":"2026-11-18T22:09:14Z","subjects":["séries temporelles, forecaster, N-Beats, LSTM, MLP, hyper-paramètres, incertitude, approximation bayésienne, intervalles de confiance, régression quantile"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["INSERT-YOUR-LIBRARY-ID-HERE2913"],"render_values":[{"text":"INSERT-YOUR-LIBRARY-ID-HERE2913","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null,"record_url":null,"record_label":null,"record_kind":null,"record_source":null,"download_url":"https://espace.etsmtl.ca/id/eprint/2913/1/YOUSEFI_Somaiieh.pdf","download_label":"Download PDF","download_source":"dc:identifier","access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://espace.etsmtl.ca/cgi/oai2?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Aespace.etsmtl.ca%3A2913","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Yousefi, Somaiieh"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2021-12-13"]},{"key":"dc:publisher","label":"Institution","values":["École de technologie supérieure"]},{"key":"dc:type","label":"Dc Type","values":["Electronic Thesis or Dissertation"]},{"key":"thesis:degree_level","label":"Degree Level","values":["masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Mémoire de maîtrise électronique"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["École de technologie supérieure"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["séries temporelles, forecaster, N-Beats, LSTM, MLP, hyper-paramètres, incertitude, approximation bayésienne, intervalles de confiance, régression quantile"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["INSERT-YOUR-LIBRARY-ID-HERE2913","https://espace.etsmtl.ca/id/eprint/2913/1/YOUSEFI_Somaiieh.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["The recent popularity of using Deep Learning models for the forecasting of time series calls for methods to not only predict the target but also to measure the uncertainty of the prediction accurately. Working with time series requires reliable forecasters. An intuitive N-Beats architecture was implemented in this thesis with the desirable properties of being fast to train and adaptable, without requiring modifications for the various time series array settings. Its performance was compared with other models, namely Long Short-Term Memory (LSTM) Multi-layer Perceptron (MLP) and a naïve method. The influence of different hyperparameters were investigated as to how the parameters affected the performances of the models. Our NBeats model outperformed well-known time series forecaster, LSTM. All the implementations were conducted in Python programming language. Random sampling was performed to avoid overfitting. Our target field data are Microsoft and NASP data sets. Estimating uncertainty is a component of reliable machine leaning models for time series forecasting. To measure the uncertainty and confidence level of the selected MLP model, Monte Carlo dropout, which approximates Bayesian uncertainty, was applied during inference. Quantile Regression was also implemented on the MLP algorithm as a baseline to predict the confidence intervals and to evaluate our strategy for estimating uncertainty. As a result, Bayesian approximation models indicated the necessity of calibration to adjust the predicted probability."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Forecasting lightpath quality of transmission and implementing uncertainty in the forecast models"]}]}],"canonical_facts":{"dc:creator":["Yousefi, Somaiieh"],"dc:date":["2021-12-13"],"dc:description":["The recent popularity of using Deep Learning models for the forecasting of time series calls for methods to not only predict the target but also to measure the uncertainty of the prediction accurately. Working with time series requires reliable forecasters. An intuitive N-Beats architecture was implemented in this thesis with the desirable properties of being fast to train and adaptable, without requiring modifications for the various time series array settings. Its performance was compared with other models, namely Long Short-Term Memory (LSTM) Multi-layer Perceptron (MLP) and a naïve method. The influence of different hyperparameters were investigated as to how the parameters affected the performances of the models. Our NBeats model outperformed well-known time series forecaster, LSTM. All the implementations were conducted in Python programming language. Random sampling was performed to avoid overfitting. Our target field data are Microsoft and NASP data sets. Estimating uncertainty is a component of reliable machine leaning models for time series forecasting. To measure the uncertainty and confidence level of the selected MLP model, Monte Carlo dropout, which approximates Bayesian uncertainty, was applied during inference. Quantile Regression was also implemented on the MLP algorithm as a baseline to predict the confidence intervals and to evaluate our strategy for estimating uncertainty. As a result, Bayesian approximation models indicated the necessity of calibration to adjust the predicted probability."],"dc:format":["application/pdf"],"dc:identifier":["INSERT-YOUR-LIBRARY-ID-HERE2913","https://espace.etsmtl.ca/id/eprint/2913/1/YOUSEFI_Somaiieh.pdf"],"dc:language":["en"],"dc:publisher":["École de technologie supérieure"],"dc:subject":["séries temporelles, forecaster, N-Beats, LSTM, MLP, hyper-paramètres, incertitude, approximation bayésienne, intervalles de confiance, régression quantile"],"dc:title":["Forecasting lightpath quality of transmission and implementing uncertainty in the forecast models"],"dc:type":["Electronic Thesis or Dissertation"],"thesis:degree_level":["masters"],"thesis:degree_name":["Mémoire de maîtrise électronique"],"thesis:institution_name":["École de technologie supérieure"]},"updated_at":"2026-11-18T22:09:14Z"}