{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T07:12:54Z","timestamp":1777705974709,"version":"3.51.4"},"reference-count":36,"publisher":"SAGE Publications","issue":"1","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IFS"],"published-print":{"date-parts":[[2022,6,1]]},"abstract":"<jats:p>Time series forecasting (TSF) is significant for many applications, therefore the exploration and study for this problem has been proceeding. With the advances of computing power, deep neural networks (DNNs) have shown powerful performance on many machine learning tasks when considerable amounts of data can be used. However, sufficient data may be unavailable in some scenarios, which leads to performance degradation or even not working of DNN-based models. In this paper, we focus on few-shot time series forecasting task and propose to employ meta-learning to alleviate the problems caused by insufficient training data. Therefore, we propose a meta-learning-based prediction mechanism for few-shot time series forecasting task, which mainly consists of meta-training and meta-testing. The meta-training phase uses first-order model-agnostic meta-learning algorithm (MAML) as a core component to conduct cross-task training, and thus our method also inherits the advantages of the MAML, i.e., model-agnostic, in the sense that our method is compatible with any model trained with gradient descent. In the meta-testing phase, the DNN-based models are fine-tuned by the small number of time series data from an unseen task in the meta-training phase. We design two groups of comparison models to validate the effectiveness of our method. The first group, as the baseline models, is trained directly on specific time series dataset from target task. The second group, as comparison models, is trained by our proposed method. Also, we conduct data sensitivity study to validate the robustness of our method. The experimental results indicate the second group models outperform the first in different degrees in terms of prediction accuracy and convergence speed, and our method has strong robustness for forecast horizons and data scales.<\/jats:p>","DOI":"10.3233\/jifs-212228","type":"journal-article","created":{"date-parts":[[2022,4,26]],"date-time":"2022-04-26T13:50:38Z","timestamp":1650981038000},"page":"325-341","source":"Crossref","is-referenced-by-count":7,"title":["Meta-learning for few-shot time series forecasting"],"prefix":"10.1177","volume":"43","author":[{"given":"Feng","family":"Xiao","sequence":"first","affiliation":[{"name":"College of Computer Science and Technology, Jilin University, Changchun, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lu","family":"Liu","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Jilin University, Changchun, China"},{"name":"College of Software, Jilin University, Changchun, China"},{"name":"Key Laboratory of Symbol Computation and Knowledge Engineering of the Ministry of Education, Changchun, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiayu","family":"Han","sequence":"additional","affiliation":[{"name":"Department of Linguistics, University of Washington, Seattle, WA, United States"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Degui","family":"Guo","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Jilin University, Changchun, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shang","family":"Wang","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Jilin University, Changchun, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hai","family":"Cui","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Jilin University, Changchun, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tao","family":"Peng","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Jilin University, Changchun, China"},{"name":"Key Laboratory of Symbol Computation and Knowledge Engineering of the Ministry of Education, Changchun, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","reference":[{"key":"10.3233\/JIFS-212228_ref1","unstructured":"Li Y. , Yu R. , Shahabi C. and Liu Y. , Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting, 6th International Conference on Learning Representations, Vancouver, May 2018."},{"key":"10.3233\/JIFS-212228_ref2","first-page":"1","article-title":"Time-series extreme event forecasting with neural networks at uber","volume":"34","author":"Laptev","year":"2017","journal-title":"International Conference on Machine Learning"},{"key":"10.3233\/JIFS-212228_ref3","doi-asserted-by":"crossref","first-page":"380","DOI":"10.1016\/j.eswa.2017.09.059","article-title":"Energy consumption forecasting based on Elman neural networks with evolutive optimization","volume":"92","author":"Ruiz","year":"2018","journal-title":"Expert Systems with Applications"},{"key":"10.3233\/JIFS-212228_ref4","doi-asserted-by":"crossref","first-page":"902","DOI":"10.1016\/j.rser.2017.02.085","article-title":"A review on time series forecasting techniques for building energy consumption","volume":"74","author":"Deb","year":"2017","journal-title":"Renewable and Sustainable Energy Reviews"},{"key":"10.3233\/JIFS-212228_ref5","first-page":"1","article-title":"EVDHM-ARIMA-Based Time Series Forecasting Model and Its Application for COVID-19 Cases","volume":"70","author":"Sharma","year":"2021","journal-title":"IEEE Transactions on Instrumentation and Measurement"},{"key":"10.3233\/JIFS-212228_ref6","doi-asserted-by":"crossref","first-page":"226","DOI":"10.1016\/j.eswa.2019.01.012","article-title":"Literature review: Machine learning techniques applied to financial market prediction","volume":"124","author":"Henrique","year":"2019","journal-title":"Expert Systems with Applications"},{"key":"10.3233\/JIFS-212228_ref7","unstructured":"Box G. and Jenkins G. , Time Series Analysis: Forecasting and Control, Journal of the American Statistical Association, 1990."},{"issue":"4","key":"10.3233\/JIFS-212228_ref8","doi-asserted-by":"crossref","first-page":"809","DOI":"10.1109\/72.935093","article-title":"Financial time series prediction using least squares support vector machines within the evidence framework","volume":"12","author":"Van Gestel","year":"2001","journal-title":"IEEE Transactions on Neural Networks"},{"key":"10.3233\/JIFS-212228_ref9","first-page":"847","article-title":"Temporal Regularized Matrix Factorization for High-dimensional Time Series Prediction","volume":"29","author":"Yu","year":"2016","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.3233\/JIFS-212228_ref10","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","volume":"25","author":"Krizhevsky","year":"2012","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.3233\/JIFS-212228_ref11","first-page":"770","article-title":"Deep Residual Learning for Image Recognition","author":"He","year":"2016","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"10.3233\/JIFS-212228_ref12","first-page":"3104","article-title":"Sequence to Sequence Learning with Neural Networks","volume":"27","author":"Sutskever","year":"2014","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.3233\/JIFS-212228_ref13","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1016\/j.neucom.2018.09.082","article-title":"Time series forecasting of petroleum production using deep LSTM recurrent networks","volume":"323","author":"Sagheer","year":"2019","journal-title":"Neurocomputing"},{"key":"10.3233\/JIFS-212228_ref14","doi-asserted-by":"crossref","first-page":"1078","DOI":"10.1016\/j.apenergy.2018.12.042","article-title":"Dayahead building-level load forecasts using deep learning vs. traditional time-series techniques","volume":"236","author":"Cai","year":"2019","journal-title":"Applied Energy"},{"issue":"4","key":"10.3233\/JIFS-212228_ref15","first-page":"73","article-title":"Dilated convolutional neural networks for time series forecasting","volume":"22","author":"Borovykh","year":"2019","journal-title":"Journal of Computational Finance"},{"issue":"1","key":"10.3233\/JIFS-212228_ref16","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1007\/BF02532251","article-title":"Fitting autoregressive models for prediction","volume":"21","author":"Akaike","year":"1969","journal-title":"Annals of the institute of Statistical Mathematics"},{"issue":"2-3","key":"10.3233\/JIFS-212228_ref17","doi-asserted-by":"crossref","first-page":"188","DOI":"10.1016\/0031-9201(76)90047-9","article-title":"Time series modelling and maximum entropy","volume":"12","author":"Ulrych","year":"1976","journal-title":"Physics of the Earth and Planetary Interiors"},{"issue":"3","key":"10.3233\/JIFS-212228_ref18","doi-asserted-by":"crossref","first-page":"346","DOI":"10.1002\/mrm.1910350312","article-title":"Movement-related effects in fMRI time-series","volume":"35","author":"Friston","year":"1996","journal-title":"Magnetic Resonance in Medicine"},{"issue":"3","key":"10.3233\/JIFS-212228_ref19","doi-asserted-by":"crossref","first-page":"1014","DOI":"10.1109\/TPWRS.2002.804943","article-title":"ARIMA models to predict next-day electricity prices","volume":"18","author":"Contreras","year":"2003","journal-title":"IEEE Transactions on Power Systems"},{"issue":"6","key":"10.3233\/JIFS-212228_ref20","doi-asserted-by":"crossref","first-page":"664","DOI":"10.1061\/(ASCE)0733-947X(2003)129:6(664)","article-title":"Modeling and forecasting vehicular traffic flow as a seasonal ARIMA process: Theoretical basis and empirical results","volume":"129","author":"Williams","year":"2003","journal-title":"Journal of Transportation Engineering"},{"key":"10.3233\/JIFS-212228_ref21","doi-asserted-by":"crossref","first-page":"159","DOI":"10.1016\/S0925-2312(01)00702-0","article-title":"Time series forecasting using a hybrid ARIMA and neural network model","volume":"50","author":"Zhang","year":"2003","journal-title":"Neurocomputing"},{"key":"10.3233\/JIFS-212228_ref22","doi-asserted-by":"crossref","first-page":"220","DOI":"10.1016\/j.econmod.2013.09.024","article-title":"Forecasting tourism demandto Catalonia: Neural networks vs. time series models","volume":"36","author":"Claveria","year":"2014","journal-title":"Economic Modelling"},{"issue":"4","key":"10.3233\/JIFS-212228_ref23","doi-asserted-by":"crossref","first-page":"802","DOI":"10.1016\/j.ijforecast.2018.06.001","article-title":"The M4 Competition: Results, findings, conclusion and way forward","volume":"34","author":"Makridakis","year":"2018","journal-title":"International Journal of Forecasting"},{"key":"10.3233\/JIFS-212228_ref24","unstructured":"Oreshkin B.N. , Carpov D. , Chapados N. and Bengio Y. , N-BEATS: Neural basis expansion analysis for interpretable time series forecasting, 8th International Conference on Learning Representations, Addis Ababa, April 2020."},{"key":"10.3233\/JIFS-212228_ref25","unstructured":"Orozco B.P. and Roberts S.J. , Zero-shot and few-shot time series forecasting with ordinal regression recurrent neural networks, 28th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Bruges, October 2020."},{"key":"10.3233\/JIFS-212228_ref26","unstructured":"Iwata T. and Kumagai A. , Few-shot Learning for Time-series Forecasting, unpublished, CoRR. Available: https:\/\/arxiv.org\/abs\/2009.14379"},{"key":"10.3233\/JIFS-212228_ref27","doi-asserted-by":"crossref","first-page":"9242","DOI":"10.1609\/aaai.v35i10.17115","article-title":"Meta-learning framework with applications to zero-shot time-series forecasting","author":"Oreshkin","year":"2021","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"10.3233\/JIFS-212228_ref28","first-page":"2554","article-title":"Meta networks","volume":"70","author":"Munkhdalai","year":"2017","journal-title":"International Conference on Machine Learning"},{"key":"10.3233\/JIFS-212228_ref29","first-page":"1126","article-title":"Model-agnostic meta-learning for fast adaptation of deep networks","volume":"70","author":"Finn","year":"2017","journal-title":"International Conference on Machine Learning"},{"key":"10.3233\/JIFS-212228_ref30","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1109\/TPAMI.2021.3079209","article-title":"Meta-learning in neural networks: A survey","author":"Hospedales","year":"2021","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"10.3233\/JIFS-212228_ref31","first-page":"4077","article-title":"Prototypical networks for few-shot learning","volume":"30","author":"Snell","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.3233\/JIFS-212228_ref32","first-page":"3630","article-title":"Matching networks for one shot learning","volume":"29","author":"Vinyals","year":"2016","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.3233\/JIFS-212228_ref33","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1142\/S0129065721300011","article-title":"An Experimental Review on Deep Learning Architectures for Time Series Forecasting","volume":"31","author":"Lara-Benitez","year":"2021","journal-title":"International Journal of Neural Systems"},{"key":"10.3233\/JIFS-212228_ref34","first-page":"1","article-title":"Convolutional Neural Networks for Energy Time Series Forecasting","author":"Koprinska","year":"2018","journal-title":"International Joint Conference on Neural Networks (IJCNN)"},{"key":"10.3233\/JIFS-212228_ref35","first-page":"267","article-title":"Very Short-Term Solar Generation Forecasting Based on LSTM with Temporal Attention Mechanism","author":"Pan","year":"2019","journal-title":"IEEE 5th International Conference on Computer and Communications (ICCC)"},{"issue":"6","key":"10.3233\/JIFS-212228_ref36","doi-asserted-by":"crossref","first-page":"1293","DOI":"10.1109\/JAS.2019.1911747","article-title":"The UCR time series archive","volume":"6","author":"Dau","year":"2019","journal-title":"{IEEE\/CAA Journal of Automatica Sinica"}],"container-title":["Journal of Intelligent &amp; Fuzzy Systems"],"original-title":[],"link":[{"URL":"https:\/\/content.iospress.com\/download?id=10.3233\/JIFS-212228","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T09:45:43Z","timestamp":1777455943000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/full\/10.3233\/JIFS-212228"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,6,1]]},"references-count":36,"journal-issue":{"issue":"1"},"URL":"https:\/\/doi.org\/10.3233\/jifs-212228","relation":{},"ISSN":["1064-1246","1875-8967"],"issn-type":[{"value":"1064-1246","type":"print"},{"value":"1875-8967","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,6,1]]}}}