{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,6]],"date-time":"2026-07-06T06:15:11Z","timestamp":1783318511853,"version":"3.54.6"},"reference-count":124,"publisher":"Springer Science and Business Media LLC","issue":"12","license":[{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T00:00:00Z","timestamp":1781568000000},"content-version":"vor","delay-in-days":15,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100005855","name":"Universidade Nova de Lisboa","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100005855","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Neural Comput &amp; Applic"],"published-print":{"date-parts":[[2026,6]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Time-series forecasting of multiple related sequences presents unique challenges due to the complex interplay between individual series characteristics and global patterns. We present\n                    <jats:italic>T2f<\/jats:italic>\n                    , a forecasting method combining ensemble learning with an actor-critic architecture based on the Twin Delayed Deep Deterministic algorithm (TD3).\n                    <jats:italic>T2f<\/jats:italic>\n                    balances local and global patterns through both its architecture and learning approaches, integrating transformer-based pattern recognition with reinforcement learning for dynamic model selection. Our method incorporates temporal attention mechanisms and context-aware error measurement, aligning forecasting objectives with practical decision-making priorities. Comprehensive ablation studies demonstrate that\n                    <jats:italic>T2f<\/jats:italic>\n                    \u2019s components provide synergistic benefits: the TD3-based optimizer contributes 18.8% error reduction over static weighting, while temporal attention adds 8.0% improvement, with the integrated system outperforming simple ensemble baselines by over 20%. Experimental results across five diverse datasets indicate\n                    <jats:italic>T2f<\/jats:italic>\n                    reduced mean absolute error by over 30% compared to statistical models and achieved up to 40% better performance on context-weighted metrics than competing approaches. While specialized models occasionally outperformed\n                    <jats:italic>T2f<\/jats:italic>\n                    on highly regular patterns, it consistently showed superior adaptability to contextual weights with faster convergence, typically reaching near-optimal performance within 25 epochs compared to 40+ for alternative methods, particularly on datasets with complex temporal dynamics.\n                  <\/jats:p>","DOI":"10.1007\/s00521-026-12209-6","type":"journal-article","created":{"date-parts":[[2026,6,16]],"date-time":"2026-06-16T01:58:27Z","timestamp":1781575107000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["T2f: Actor-critic reinforcement learning for time-series forecasting"],"prefix":"10.1007","volume":"38","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5600-9707","authenticated-orcid":false,"given":"Jo\u00e3o","family":"Sousa","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Roberto","family":"Henriques","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,6,16]]},"reference":[{"key":"12209_CR1","unstructured":"Seeger M, Salinas D, Flunkert V (2016) Bayesian intermittent demand forecasting for large inventories. Advances in Neural Information Processing Systems. pp 4646\u20134654"},{"key":"12209_CR2","unstructured":"Rangapuram SS, Seeger M, Gasthaus J, Stella L, Wang Y, Januschowski T (2018) Deep state space models for time series forecasting. Neural Information Processing Systems. pp 7796\u20137805"},{"key":"12209_CR3","doi-asserted-by":"crossref","DOI":"10.1515\/9780691218632","volume-title":"Time series analysis","author":"JD Hamilton","year":"1994","unstructured":"Hamilton JD (1994) Time series analysis. Princeton University Press"},{"key":"12209_CR4","volume-title":"Time-Series Forecasting","author":"C Chatfield","year":"2000","unstructured":"Chatfield C (2000) Time-Series Forecasting. Chapman & Hall\/CRC, Boca Raton, FL"},{"issue":"3","key":"12209_CR5","doi-asserted-by":"crossref","first-page":"333","DOI":"10.1002\/for.3980030312","volume":"3","author":"E McKenzie","year":"1984","unstructured":"McKenzie E (1984) General exponential smoothing and the equivalent ARMA process. J Forecast 3(3):333\u2013344","journal-title":"J Forecast"},{"issue":"12","key":"12209_CR6","doi-asserted-by":"crossref","first-page":"1694","DOI":"10.14778\/3137765.3137775","volume":"10","author":"V Flunkert","year":"2017","unstructured":"Flunkert V, Gasthaus J, Januschowski T, Lange D, Salinas D, Schelter S, Seeger M, Wang Y (2017) Probabilistic demand forecasting at scale. Proceedings of the VLDB Endowment 10(12):1694\u20131705","journal-title":"Proceedings of the VLDB Endowment"},{"issue":"12","key":"12209_CR7","doi-asserted-by":"crossref","first-page":"2102","DOI":"10.14778\/3229863.3229878","volume":"11","author":"C Faloutsos","year":"2018","unstructured":"Faloutsos C, Gasthaus J, Januschowski T, Wang Y (2018) Forecasting big time series: Old and new. Proceedings of the VLDB Endowment 11(12):2102\u20132105","journal-title":"Proceedings of the VLDB Endowment"},{"key":"12209_CR8","unstructured":"Gasthaus J, Benidis K, Wang Y, Rangapuram SS, Salinas D, Flunkert V, Januschowski T (2019) Probabilistic forecasting with spline quantile function RNNs. Neural Information Processing Systems"},{"issue":"3","key":"12209_CR9","doi-asserted-by":"crossref","first-page":"1181","DOI":"10.1016\/j.ijforecast.2019.07.001","volume":"36","author":"D Salinas","year":"2020","unstructured":"Salinas D, Flunkert V, Gasthaus J, Januschowski T (2020) Deepar: Probabilistic forecasting with autoregressive recurrent networks. Int J Forecast 36(3):1181\u20131191","journal-title":"Int J Forecast"},{"key":"12209_CR10","doi-asserted-by":"publisher","first-page":"221","DOI":"10.1016\/j.neucom.2022.08.007","volume":"507","author":"F Lin","year":"2022","unstructured":"Lin F, Xu Y, Zhang Z, Gao C, Yamada KD (2022) Cosmos propagation network: Deep learning model for point cloud completion. Neurocomputing 507:221\u2013234. https:\/\/doi.org\/10.1016\/j.neucom.2022.08.007","journal-title":"Neurocomputing"},{"key":"12209_CR11","doi-asserted-by":"publisher","first-page":"5765","DOI":"10.1109\/TIP.2022.3205216","volume":"31","author":"A Jain","year":"2022","unstructured":"Jain A, Shaker S (2022) A data augmentation pipeline to generate synthetic labeled datasets of 3d echocardiography images using a gan. IEEE Trans Image Process 31:5765\u20135776. https:\/\/doi.org\/10.1109\/TIP.2022.3205216","journal-title":"IEEE Trans Image Process"},{"key":"12209_CR12","doi-asserted-by":"publisher","DOI":"10.1016\/j.trc.2024.104838","volume":"167","author":"Y Chen","year":"2024","unstructured":"Chen Y, Zheng L, Tan Z (2024) Roadside lidar placement for cooperative traffic detection by a novel chance constrained stochastic simulation optimization approach. Transportation Research Part C Emerging Technologies 167:104838. https:\/\/doi.org\/10.1016\/j.trc.2024.104838","journal-title":"Transportation Research Part C Emerging Technologies"},{"key":"12209_CR13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s40747-024-01693-9","volume":"11","author":"Y Chen","year":"2024","unstructured":"Chen Y et al (2024) Transforming traffic accident investigations: a virtual-real-fusion framework for intelligent 3d traffic accident reconstruction. Complex Intell Syst 11:1\u201318. https:\/\/doi.org\/10.1007\/s40747-024-01693-9","journal-title":"Complex Intell Syst"},{"key":"12209_CR14","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2025.114514","volume":"313","author":"Y Chen","year":"2025","unstructured":"Chen Y et al (2025) Energy-efficient adaptive perception for autonomous driving via lightweight policy learning and simulation-based optimization. Knowl-Based Syst 313:114514. https:\/\/doi.org\/10.1016\/j.knosys.2025.114514","journal-title":"Knowl-Based Syst"},{"key":"12209_CR15","unstructured":"Mukherjee S, Shankar D, Ghosh A, Tathawadekar N, Kompalli P, Sarawagi S, Chaudhury K (2018) AR-MDN: Associative and recurrent mixture density networks for eRetail demand forecasting. Neural Information Processing Systems"},{"key":"12209_CR16","unstructured":"Bai S, Kolter JZ, Koltun V (2018) An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv preprint arXiv:1803.01271"},{"key":"12209_CR17","doi-asserted-by":"crossref","unstructured":"Li S, Jin X, Xuan Y, Zhou X, Chen W, Wang Y-X, Yan X (2020) Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting. Adv Neural Inf Process Syst 32:","DOI":"10.1109\/TNNLS.2020.2980749"},{"issue":"1","key":"12209_CR18","first-page":"28","volume":"39","author":"F Petropoulos","year":"2023","unstructured":"Petropoulos F, Nikolopoulos K (2023) Model selection in time series forecasting: The case for dynamic combination. Int J Forecast 39(1):28\u201343","journal-title":"Int J Forecast"},{"key":"12209_CR19","doi-asserted-by":"crossref","unstructured":"Feng C, Zhang J (2018) Reinforcement learning based dynamic model selection for short-term load forecasting. IEEE Transactions on Smart Grid","DOI":"10.1109\/ISGT.2019.8791671"},{"key":"12209_CR20","doi-asserted-by":"crossref","unstructured":"Yuan F, Shou L, Pei J, Lin W, Gong M, Fu Y, Jiang D (2021) Reinforced multi-teacher selection for knowledge distillation. AAAI Conference on Artificial Intelligence","DOI":"10.1609\/aaai.v35i16.17680"},{"key":"12209_CR21","volume":"208","author":"T Liu","year":"2020","unstructured":"Liu T, Tan Z, Xu C, Chen H, Li Z (2020) Study on deep reinforcement learning techniques for building energy consumption forecasting. Energy and Buildings 208:109675","journal-title":"Energy and Buildings"},{"issue":"12","key":"12209_CR22","first-page":"1","volume":"55","author":"R Wen","year":"2023","unstructured":"Wen R, Torkkola K, Narayanaswamy B (2023) A survey on multi-step time series forecasting with deep learning. ACM Comput Surv 55(12):1\u201338","journal-title":"ACM Comput Surv"},{"key":"12209_CR23","volume-title":"Bayesian Data Analysis","author":"A Gelman","year":"2004","unstructured":"Gelman A, Stern HS, Carlin JB, Rubin DB (2004) Bayesian Data Analysis. Routledge, New York"},{"key":"12209_CR24","doi-asserted-by":"crossref","unstructured":"Ahmed A, Aly M, Gonzalez J, Narayanamurthy S, Smola A (2012) Scalable inference in latent variable models general terms. Proceedings of the Fifth ACM International Conference on Web Search and Data Mining","DOI":"10.1145\/2124295.2124312"},{"issue":"4","key":"12209_CR25","doi-asserted-by":"crossref","first-page":"540","DOI":"10.1162\/003465300559037","volume":"82","author":"M Forni","year":"2000","unstructured":"Forni M, Hallin M, Lippi M, Reichlin L (2000) The generalized dynamic-factor model: identification and estimation. Rev Econ Stat 82(4):540\u2013554","journal-title":"Rev Econ Stat"},{"key":"12209_CR26","doi-asserted-by":"crossref","first-page":"1341","DOI":"10.1162\/neco.1996.8.7.1341","volume":"8","author":"DH Wolpert","year":"1996","unstructured":"Wolpert DH (1996) The lack of a priori distinctions between learning algorithms. Neural Comput 8:1341\u20131390","journal-title":"Neural Comput"},{"key":"12209_CR27","unstructured":"Fujimoto S, Van Hoof H, Meger D (2018) Addressing function approximation error in actor-critic methods. International Conference on Machine Learning"},{"key":"12209_CR28","doi-asserted-by":"crossref","unstructured":"Montero-Manso P, Athanasopoulos G, Hyndman RJ, Talagala TS (2019) FFORMA: Feature-based forecast model averaging. Int J Forecast","DOI":"10.1016\/j.ijforecast.2019.02.011"},{"issue":"2194","key":"12209_CR29","volume":"379","author":"B Lim","year":"2021","unstructured":"Lim B, Zohren S (2021) Time series forecasting with deep learning: a survey. Philos Trans R Soc A Math Phys Eng Sci 379(2194):20200209","journal-title":"Philos Trans R Soc A Math Phys Eng Sci"},{"key":"12209_CR30","doi-asserted-by":"crossref","unstructured":"Qin Y, Song D, Chen H, Cheng W, Jiang G, Cottrell G (2017) Dual-stage attention-based recurrent neural network for time series prediction. In: Proceedings of the 26th International Joint Conference on Artificial Intelligence, pp. 2627\u20132633","DOI":"10.24963\/ijcai.2017\/366"},{"key":"12209_CR31","unstructured":"Li S, Jin X, Xuan Y, Zhou X, Chen W, Wang Y-X, Yan X (2019) Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting. Advances in Neural Information Processing Systems. pp 5243\u20135253"},{"key":"12209_CR32","unstructured":"Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser L, Polosukhin I (2017) Attention is all you need. Advances in Neural Information Processing Systems. pp 5998\u20136008"},{"key":"12209_CR33","doi-asserted-by":"crossref","first-page":"1421","DOI":"10.1007\/s10994-019-05815-0","volume":"108","author":"S-Y Shih","year":"2019","unstructured":"Shih S-Y, Sun F-K, Lee H-Y (2019) Temporal pattern attention for multivariate time series forecasting. Machine Learning 108:1421\u20131441","journal-title":"Machine Learning"},{"key":"12209_CR34","first-page":"216","volume":"28","author":"DT Kim","year":"2017","unstructured":"Kim DT, Ahn HJ, Choi E (2017) Temporal attention augmented bilinear network for financial time-series data analysis. IEEE Transactions on Neural Networks and Learning Systems 28:216\u2013227","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"12209_CR35","doi-asserted-by":"publisher","DOI":"10.3389\/frai.2022.924688","volume":"5","author":"KD Yamada","year":"2022","unstructured":"Yamada KD, Baladram MS, Lin F (2022) Relation is an option for processing context information. Frontiers in Artificial Intelligence 5:924688. https:\/\/doi.org\/10.3389\/frai.2022.924688","journal-title":"Frontiers in Artificial Intelligence"},{"issue":"10","key":"12209_CR36","doi-asserted-by":"crossref","first-page":"2451","DOI":"10.1162\/089976600300015015","volume":"12","author":"FA Gers","year":"1999","unstructured":"Gers FA, Schmidhuber J, Cummins F (1999) Learning to forget: Continual prediction with lstm. Neural Comput 12(10):2451\u20132471","journal-title":"Neural Comput"},{"key":"12209_CR37","unstructured":"Chung J, Gulcehre C, Cho K, Bengio Y (2014) Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling. arxiv.org\/abs\/1412.3555"},{"key":"12209_CR38","unstructured":"Yu F, Koltun V (2016) Multi-scale context aggregation by dilated convolutions. International Conference on Learning Representations. arxiv.org\/abs\/1511.07122"},{"key":"12209_CR39","unstructured":"Thrun S, Schwartz A (1993) Issues in using function approximation for reinforcement learning. Proceedings of the 4th Connectionist Models Summer School. pp 255\u2013263"},{"issue":"9","key":"12209_CR40","doi-asserted-by":"crossref","first-page":"1150","DOI":"10.1057\/palgrave.jors.2602597","volume":"59","author":"R Fildes","year":"2008","unstructured":"Fildes R, Nikolopoulos K, Crone SF, Syntetos AA (2008) Forecasting and operational research: a review. Journal of the Operational Research Society 59(9):1150\u20131172","journal-title":"Journal of the Operational Research Society"},{"key":"12209_CR41","first-page":"131","volume":"193","author":"X Wan","year":"2017","unstructured":"Wan X, Evers PT, Dresner ME (2017) Retail inventory management with stock-out based dynamic demand substitution. Int J Prod Econ 193:131\u2013141","journal-title":"Int J Prod Econ"},{"issue":"1","key":"12209_CR42","doi-asserted-by":"crossref","first-page":"207","DOI":"10.32614\/RJ-2017-009","volume":"9","author":"S Moritz","year":"2017","unstructured":"Moritz S, Bartz-Beielstein T (2017) Imputets: time series missing value imputation in r. The R Journal 9(1):207","journal-title":"The R Journal"},{"key":"12209_CR43","doi-asserted-by":"publisher","unstructured":"Tawakuli A, Havers B, Gulisano V, Kaiser D, Engel T (2024) Survey: Time-series data preprocessing: A survey and an empirical analysis. Journal of Engineering Research. https:\/\/doi.org\/10.1016\/j.jer.2024.02.018","DOI":"10.1016\/j.jer.2024.02.018"},{"issue":"460","key":"12209_CR44","doi-asserted-by":"crossref","first-page":"1167","DOI":"10.1198\/016214502388618960","volume":"97","author":"JH Stock","year":"2002","unstructured":"Stock JH, Watson MW (2002) Forecasting using principal components from a large number of predictors. J Am Stat Assoc 97(460):1167\u20131179","journal-title":"J Am Stat Assoc"},{"issue":"471","key":"12209_CR45","doi-asserted-by":"crossref","first-page":"830","DOI":"10.1198\/016214504000002050","volume":"100","author":"M Forni","year":"2005","unstructured":"Forni M, Hallin M, Lippi M, Reichlin L (2005) The generalized dynamic factor model: one-sided estimation and forecasting. J Am Stat Assoc 100(471):830\u2013840","journal-title":"J Am Stat Assoc"},{"issue":"2","key":"12209_CR46","doi-asserted-by":"crossref","first-page":"147","DOI":"10.1198\/073500102317351921","volume":"20","author":"JH Stock","year":"2002","unstructured":"Stock JH, Watson MW (2002) Macroeconomic forecasting using diffusion indexes. Journal of Business & Economic Statistics 20(2):147\u2013162","journal-title":"Journal of Business & Economic Statistics"},{"key":"12209_CR47","volume-title":"Time Series Analysis: Forecasting and Control","author":"GE Box","year":"2015","unstructured":"Box GE, Jenkins GM, Reinsel GC, Ljung GM (2015) Time Series Analysis: Forecasting and Control. John Wiley & Sons, Hoboken, NJ"},{"key":"12209_CR48","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-540-71918-2","volume-title":"Forecasting with Exponential Smoothing: the State Space Approach","author":"R Hyndman","year":"2008","unstructured":"Hyndman R, Koehler AB, Ord JK, Snyder RD (2008) Forecasting with Exponential Smoothing: the State Space Approach. Springer, New York"},{"key":"12209_CR49","doi-asserted-by":"crossref","DOI":"10.1093\/acprof:oso\/9780199641178.001.0001","volume-title":"Time Series Analysis by State Space Methods","author":"J Durbin","year":"2012","unstructured":"Durbin J, Koopman SJ (2012) Time Series Analysis by State Space Methods. Oxford University Press, Oxford, UK"},{"key":"12209_CR50","unstructured":"Sen R, Yu H-F, Dhillon I (2019) Think globally, act locally: A deep neural network approach to high-dimensional time series forecasting. Advances in Neural Information Processing Systems"},{"key":"12209_CR51","unstructured":"Wang Y, Smola A, Maddix D, Gasthaus J, Foster D, Januschowski T (2019) Deep factors for forecasting. International Conference on Machine Learning"},{"key":"12209_CR52","unstructured":"Tang ZR, Zhuang F, Chi W, Liu M, Chen X (2020) Memory-augmented neural networks for predictive process analytics. IEEE Trans Knowl Data Eng"},{"key":"12209_CR53","doi-asserted-by":"crossref","unstructured":"Zhou H, Zhang S, Peng J, Zhang S, Li J, Xiong H, Zhang W (2021) Informer: Beyond efficient transformer for long sequence time-series forecasting. In: Proceedings of AAAI","DOI":"10.1609\/aaai.v35i12.17325"},{"key":"12209_CR54","first-page":"22419","volume":"34","author":"H Wu","year":"2021","unstructured":"Wu H, Xu J, Wang J, Long M (2021) Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting. Adv Neural Inf Process Syst 34:22419\u201322430","journal-title":"Adv Neural Inf Process Syst"},{"key":"12209_CR55","unstructured":"Zhou T, Ma Z, Wen Q, Wang X, Sun L, Jin R (2022) Fedformer: Frequency enhanced decomposed transformer for long-term series forecasting. International Conference on Machine Learning. pp 41133\u201341155"},{"key":"12209_CR56","unstructured":"Nie Y, Nguyen NH, Sinthong P, Kalagnanam J (2023) A time series is worth 64 words: Long-term forecasting with transformers. International Conference on Learning Representations"},{"key":"12209_CR57","unstructured":"Liu Y, Hu T, Zhang H, Wu H, Wang S, Ma L, Long M (2024) itransformer: Inverted transformers are effective for time series forecasting. International Conference on Learning Representations.(Spotlight)"},{"key":"12209_CR58","unstructured":"Wang S, Wu H, Shi X, Hu T, Luo H, Ma L, Zhang JY, Zhou J (2024) Timemixer: Decomposable multiscale mixing for time series forecasting. International Conference on Learning Representations"},{"key":"12209_CR59","doi-asserted-by":"crossref","unstructured":"Wang Y, Wu H, Dong J, Liu Y, Qiu Y, Zhang H, Wang J, Long M (2024) Timexer: Empowering transformers for time series forecasting with exogenous variables. In: Advances in Neural Information Processing Systems","DOI":"10.52202\/079017-0015"},{"key":"12209_CR60","unstructured":"Ansari AF, Stella L, Turkmen C, Zhang X, Mercado P, Shen H, Shchur O, Rangapuram SS, Arber SP, Kapoor S (2024) Chronos: Learning the language of time series. Transactions on Machine Learning Research"},{"key":"12209_CR61","unstructured":"Woo G, Liu C, Kumar A, Xiong C, Savarese S, Sahoo D (2024) Unified training of universal time series forecasting transformers. International Conference on Machine Learning"},{"key":"12209_CR62","unstructured":"Das A, Kong W, Leber A, Mathur M, Ope R, Sen S, Yu R (2024) A decoder-only foundation model for time-series forecasting. International Conference on Machine Learning"},{"key":"12209_CR63","first-page":"441","volume":"17","author":"J Moody","year":"1998","unstructured":"Moody J, Saffell M (1998) Performance functions and reinforcement learning for trading systems and portfolios. Journal of Forecasting Special Issue on Application of Artificial Neural Networks to Forecasting Financial Markets 17:441\u2013470","journal-title":"Journal of Forecasting Special Issue on Application of Artificial Neural Networks to Forecasting Financial Markets"},{"issue":"1","key":"12209_CR64","first-page":"87","volume":"1","author":"B Hambly","year":"2009","unstructured":"Hambly B, Jones O, Thirkettle M, Walker B (2009) Modelling spike trains with markov chains. Journal of Mathematical Neuroscience 1(1):87\u201399","journal-title":"Journal of Mathematical Neuroscience"},{"key":"12209_CR65","unstructured":"Jiang Z, Xu D, Liang J (2017) Deep reinforcement learning for portfolio management. arXiv preprint arXiv:1706.10059"},{"issue":"3","key":"12209_CR66","doi-asserted-by":"crossref","first-page":"653","DOI":"10.1109\/TNNLS.2016.2522401","volume":"28","author":"Y Deng","year":"2016","unstructured":"Deng Y, Bao F, Kong Y, Ren Z, Dai Q (2016) Deep direct reinforcement learning for financial signal representation and trading. IEEE Transactions on Neural Networks and Learning Systems 28(3):653\u2013664","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"12209_CR67","unstructured":"Feng Y, Zhang R (2018) Dynamic model selection for time series forecasting. IEEE International Conference on Data Mining"},{"key":"12209_CR68","doi-asserted-by":"crossref","unstructured":"Qiu X, Zhang L, Ren Y, Suganthan PN, Amaratunga G (2014) Ensemble deep learning for regression and time series forecasting. IEEE Symposium on Computational Intelligence in Ensemble Learning. pp 1\u20136","DOI":"10.1109\/CIEL.2014.7015739"},{"issue":"1","key":"12209_CR69","first-page":"79","volume":"55","author":"Y Baram","year":"2004","unstructured":"Baram Y, El-Yaniv R, Luz K (2004) Mdp-based expert selection. Mach Learn 55(1):79\u2013107","journal-title":"Mach Learn"},{"key":"12209_CR70","unstructured":"Konda VR, Tsitsiklis JN (1999) Actor-critic algorithms. Advances in Neural Information Processing Systems. pp 1008\u20131014"},{"key":"12209_CR71","volume-title":"Reinforcement Learning: An Introduction","author":"RS Sutton","year":"2018","unstructured":"Sutton RS, Barto AG (2018) Reinforcement Learning: An Introduction. MIT Press, Cambridge, MA"},{"key":"12209_CR72","unstructured":"Silver D, Lever G, Heess N, Degris T, Wierstra D, Riedmiller M (2014) Deterministic policy gradient algorithms. International Conference on Machine Learning. pp 387\u2013395"},{"key":"12209_CR73","unstructured":"Lillicrap TP, Hunt JJ, Pritzel A, Heess N, Erez T, Tassa Y, Silver D, Wierstra D (2015) Continuous control with deep reinforcement learning. International Conference on Learning Representations"},{"key":"12209_CR74","unstructured":"Fu H, Li C, Liu X, Gao J, Celikyilmaz A, Carin L (2022) Reinforcement learning for model combination in time series forecasting. arXiv preprint arXiv:2203.05924"},{"key":"12209_CR75","unstructured":"Thrun S, Schwartz A (1993) Issues in using function approximation for reinforcement learning. Proceedings of the 1993 Connectionist Models Summer School"},{"issue":"4","key":"12209_CR76","first-page":"905","volume":"29","author":"S Mannor","year":"2004","unstructured":"Mannor S, Simester D, Sun P, Tsitsiklis JN (2004) The bias of q-learning. Math Oper Res 29(4):905\u2013931","journal-title":"Math Oper Res"},{"key":"12209_CR77","unstructured":"Haarnoja T, Zhou A, Abbeel P, Levine S (2018) Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor. International Conference on Machine Learning. pp 1861\u20131870 (PMLR)"},{"key":"12209_CR78","unstructured":"Fujimoto S, Meger D, Precup D (2019) Off-policy deep reinforcement learning without exploration. International Conference on Machine Learning. pp 2052\u20132062 (PMLR)"},{"key":"12209_CR79","doi-asserted-by":"crossref","unstructured":"Haarnoja T, Ha S, Zhou A, Tan J, Tucker G, Levine S (2018) Learning to walk via deep reinforcement learning.arXiv preprint arXiv:1812.11103","DOI":"10.15607\/RSS.2019.XV.011"},{"key":"12209_CR80","unstructured":"Tessler C, Caspi I, Mannor S (2019) Distributional multivariate policy evaluation and exploration with the bellman gan. International Conference on Machine Learning. pp 6167\u20136176 (PMLR)"},{"issue":"6","key":"12209_CR81","doi-asserted-by":"crossref","first-page":"4909","DOI":"10.1109\/TITS.2021.3054625","volume":"23","author":"BR Kiran","year":"2021","unstructured":"Kiran BR, Sobh I, Talpaert V, Mannion P, Al Sallab AA, Yogamani S, Perez P (2021) Deep reinforcement learning for autonomous driving: a survey. IEEE Trans Intell Transp Syst 23(6):4909\u20134926","journal-title":"IEEE Trans Intell Transp Syst"},{"key":"12209_CR82","doi-asserted-by":"crossref","first-page":"6639","DOI":"10.1609\/aaai.v36i6.20618","volume":"36","author":"H Fu","year":"2022","unstructured":"Fu H, Li C, Liu X, Gao J, Celikyilmaz A, Carin L (2022) Reinforcement learning based dynamic model combination for time series forecasting. Proceedings of the AAAI Conference on Artificial Intelligence 36:6639\u20136647","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"12209_CR83","volume-title":"Dynamic ensemble for probabilistic time-series forecasting via deep reinforcement learning","author":"I Koprinska","year":"2024","unstructured":"Koprinska I, Rana M (2024) Dynamic ensemble for probabilistic time-series forecasting via deep reinforcement learning. Amazon Science"},{"key":"12209_CR84","volume":"1","author":"Y He","year":"2024","unstructured":"He Y, Li H, Wang S (2024) Yao S (2024) A novel ensemble deep reinforcement learning model for short-term load forecasting based on q-learning dynamic model selection. The Journal of Engineering 1:e12409","journal-title":"The Journal of Engineering"},{"key":"12209_CR85","first-page":"1","volume":"57","author":"JF Torres","year":"2024","unstructured":"Torres JF, Hadjout D, Sebaa A, Martinez-Alvarez F, Troncoso A (2024) Deep learning-based time series forecasting. Artif Intell Rev 57:1\u201350","journal-title":"Artif Intell Rev"},{"key":"12209_CR86","doi-asserted-by":"crossref","unstructured":"Wen Q, Zhou T, Zhang C, Chen W, Ma Z, Yan J, Sun L (2023) Transformers in time series: A survey. International Joint Conference on Artificial Intelligence","DOI":"10.24963\/ijcai.2023\/759"},{"key":"12209_CR87","unstructured":"Yuan H, Li H, Yu H, He Z, Li G (2021) Semi-supervised deep reinforcement learning for travel demand prediction. IEEE Transactions on Neural Networks and Learning Systems"},{"issue":"4","key":"12209_CR88","doi-asserted-by":"crossref","first-page":"1748","DOI":"10.1016\/j.ijforecast.2021.03.012","volume":"37","author":"B Lim","year":"2021","unstructured":"Lim B, Ar\u00edk SO, Loeff N, Pfister T (2021) Temporal fusion transformers for interpretable multi-horizon time series forecasting. Int J Forecast 37(4):1748\u20131764","journal-title":"Int J Forecast"},{"issue":"4","key":"12209_CR89","doi-asserted-by":"crossref","first-page":"451","DOI":"10.1057\/jors.1969.103","volume":"20","author":"JM Bates","year":"1969","unstructured":"Bates JM, Granger CW (1969) The combination of forecasts. Journal of the Operational Research Society 20(4):451\u2013468","journal-title":"Journal of the Operational Research Society"},{"issue":"4","key":"12209_CR90","doi-asserted-by":"crossref","first-page":"802","DOI":"10.1016\/j.ijforecast.2018.06.001","volume":"34","author":"S Makridakis","year":"2018","unstructured":"Makridakis S, Spiliotis E, Assimakopoulos V (2018) The m4 competition: Results, findings, conclusion and way forward. Int J Forecast 34(4):802\u2013808","journal-title":"Int J Forecast"},{"key":"12209_CR91","doi-asserted-by":"crossref","unstructured":"Makridakis S, Spiliotis E, Assimakopoulos V (2022) The m5 accuracy competition: Results, findings and conclusions. Int J Forecast","DOI":"10.1016\/j.ijforecast.2021.11.013"},{"key":"12209_CR92","doi-asserted-by":"crossref","first-page":"135","DOI":"10.1016\/S1574-0706(05)01004-9","volume":"1","author":"A Timmermann","year":"2006","unstructured":"Timmermann A (2006) Forecast combinations. Handbook of Economic Forecasting 1:135\u2013196","journal-title":"Handbook of Economic Forecasting"},{"key":"12209_CR93","doi-asserted-by":"crossref","first-page":"233","DOI":"10.1146\/annurev-statistics-041715-033413","volume":"3","author":"G Claeskens","year":"2016","unstructured":"Claeskens G, Magnus JR, Vasnev AL, Wang W (2016) Statistical model choice. Annu Rev Stat Appl 3:233\u2013256","journal-title":"Annu Rev Stat Appl"},{"issue":"4","key":"12209_CR94","doi-asserted-by":"crossref","first-page":"559","DOI":"10.1016\/0169-2070(89)90012-5","volume":"5","author":"RT Clemen","year":"1989","unstructured":"Clemen RT (1989) Combining forecasts: A review and annotated bibliography. Int J Forecast 5(4):559\u2013583","journal-title":"Int J Forecast"},{"issue":"2","key":"12209_CR95","doi-asserted-by":"crossref","first-page":"197","DOI":"10.1002\/for.3980030207","volume":"3","author":"CW Granger","year":"1984","unstructured":"Granger CW, Ramanathan R (1984) Improved methods of combining forecasts. J Forecast 3(2):197\u2013204","journal-title":"J Forecast"},{"issue":"4","key":"12209_CR96","doi-asserted-by":"crossref","first-page":"382","DOI":"10.1214\/ss\/1009212519","volume":"14","author":"JA Hoeting","year":"1999","unstructured":"Hoeting JA, Madigan D, Raftery AE, Volinsky CT (1999) Bayesian model averaging: a tutorial. Stat Sci 14(4):382\u2013401","journal-title":"Stat Sci"},{"issue":"5","key":"12209_CR97","doi-asserted-by":"crossref","first-page":"1155","DOI":"10.1175\/MWR2906.1","volume":"133","author":"AE Raftery","year":"2005","unstructured":"Raftery AE, Gneiting T, Balabdaoui F, Polakowski M (2005) Using bayesian model averaging to calibrate forecast ensembles. Mon Weather Rev 133(5):1155\u20131174","journal-title":"Mon Weather Rev"},{"key":"12209_CR98","doi-asserted-by":"crossref","unstructured":"Freund Y, Schapire RE (1997) A decision-theoretic generalization of on-line learning and an application to boosting. European Conference on Computational Learning Theory. Springer, p 23\u201337","DOI":"10.1007\/3-540-59119-2_166"},{"key":"12209_CR99","first-page":"2339","volume":"13","author":"S Arora","year":"2012","unstructured":"Arora S, Hazan E, Kale S (2012) Multiplicative weights hedge in linear time. J Mach Learn Res 13:2339\u20132364","journal-title":"J Mach Learn Res"},{"key":"12209_CR100","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1561\/2200000018","volume":"4","author":"S Shalev-Shwartz","year":"2012","unstructured":"Shalev-Shwartz S (2012) Online learning and online convex optimization. Found Trends Mach Learn 4:107\u2013194","journal-title":"Found Trends Mach Learn"},{"issue":"4","key":"12209_CR101","doi-asserted-by":"crossref","first-page":"1632","DOI":"10.1016\/j.ijforecast.2021.03.004","volume":"37","author":"P Montero-Manso","year":"2021","unstructured":"Montero-Manso P, Hyndman RJ (2021) Principles and algorithms for forecasting groups of time series: Locality and globality. Int J Forecast 37(4):1632\u20131653","journal-title":"Int J Forecast"},{"key":"12209_CR102","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2021.107857","volume":"112","author":"A Incremona","year":"2021","unstructured":"Incremona A, De Nicolao G, Fusco F, Eck BJ, Tirupathi S (2021) Aggregation of nonlinearly enhanced experts with application to electricity load forecasting. Appl Soft Comput 112:107857. https:\/\/doi.org\/10.1016\/j.asoc.2021.107857","journal-title":"Appl Soft Comput"},{"issue":"1","key":"12209_CR103","doi-asserted-by":"crossref","first-page":"75","DOI":"10.1016\/j.ijforecast.2019.03.017","volume":"36","author":"S Smyl","year":"2020","unstructured":"Smyl S (2020) A hybrid method of exponential smoothing and recurrent neural networks for time series forecasting. Int J Forecast 36(1):75\u201385","journal-title":"Int J Forecast"},{"key":"12209_CR104","unstructured":"Oreshkin BN, Carpov D, Chapados N, Bengio Y (2020) Meta-learning framework with applications to zero-shot time-series forecasting. CoRR abs\/2002.02887"},{"issue":"4","key":"12209_CR105","doi-asserted-by":"crossref","first-page":"679","DOI":"10.1016\/j.ijforecast.2006.03.001","volume":"22","author":"RJ Hyndman","year":"2006","unstructured":"Hyndman RJ, Koehler AB (2006) Another look at measures of forecast accuracy. Int J Forecast 22(4):679\u2013688","journal-title":"Int J Forecast"},{"issue":"3","key":"12209_CR106","doi-asserted-by":"crossref","first-page":"788","DOI":"10.1016\/j.ijforecast.2015.12.004","volume":"32","author":"S Kolassa","year":"2016","unstructured":"Kolassa S (2016) Evaluating predictive count data distributions in retail sales forecasting. Int J Forecast 32(3):788\u2013803","journal-title":"Int J Forecast"},{"issue":"1","key":"12209_CR107","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1016\/j.ijforecast.2019.05.008","volume":"36","author":"T Januschowski","year":"2020","unstructured":"Januschowski T, Gasthaus J, Wang Y, Salinas D, Flunkert V, Bohlke-Schneider M, Callot L (2020) Criteria for classifying forecasting methods. Int J Forecast 36(1):167\u2013177","journal-title":"Int J Forecast"},{"key":"12209_CR108","doi-asserted-by":"crossref","unstructured":"Faloutsos C, Flunkert V, Gasthaus J, Januschowski T, Wang Y (2019) Classical time series forecasting methods and their applications to big data analysis: Tutorial. International Workshop on Big Time Series Analysis. pp 41\u201342","DOI":"10.1145\/3292500.3332289"},{"key":"12209_CR109","unstructured":"Kuznetsov V, Mariet Z (2020) Foundations of sequence-to-sequence modeling for time series. Transactions on Machine Learning Research"},{"key":"12209_CR110","doi-asserted-by":"crossref","DOI":"10.1017\/CBO9780511804779","volume-title":"Bayesian reasoning and machine learning","author":"D Barber","year":"2012","unstructured":"Barber D (2012) Bayesian reasoning and machine learning. Cambridge University Press"},{"key":"12209_CR111","doi-asserted-by":"crossref","unstructured":"Van Hasselt H, Guez A, Silver D (2015) Deep reinforcement learning with double Q-learning. AAAI Conference on Artificial Intelligence","DOI":"10.1609\/aaai.v30i1.10295"},{"issue":"4","key":"12209_CR112","first-page":"1","volume":"14","author":"H Tang","year":"2020","unstructured":"Tang H, Peng H, Wang Y (2020) Memory augmented neural networks for time series forecasting. ACM Trans Knowl Discov Data 14(4):1\u201327","journal-title":"ACM Trans Knowl Discov Data"},{"key":"12209_CR113","unstructured":"Oreshkin BN, Carpov D, Chapados N, Bengio Y (2020) N-beats: Neural basis expansion analysis for interpretable time series forecasting. In: International Conference on Learning Representations . https:\/\/openreview.net\/forum?id=r1ecqn4YwB"},{"key":"12209_CR114","doi-asserted-by":"publisher","DOI":"10.24432\/C58K54","author":"G Hebrail","year":"2006","unstructured":"Hebrail G, Berard A (2006) Individual Household Electric Power Consumption. UCI Machine Learning Repository. https:\/\/doi.org\/10.24432\/C58K54","journal-title":"UCI Machine Learning Repository"},{"key":"12209_CR115","doi-asserted-by":"publisher","unstructured":"Trindade A (2015) ElectricityLoadDiagrams20112014. UCI Machine Learning Repository. https:\/\/doi.org\/10.24432\/C58C86","DOI":"10.24432\/C58C86"},{"key":"12209_CR116","unstructured":"Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L (2019) Pytorch: An imperative style, high-performance deep learning library. Adv Neural Inf Process Syst 32:"},{"issue":"3","key":"12209_CR117","first-page":"1","volume":"26","author":"RJ Hyndman","year":"2008","unstructured":"Hyndman RJ, Khandakar Y (2008) Automatic time series forecasting: the forecast package for r. J Stat Softw 26(3):1\u201322","journal-title":"J Stat Softw"},{"issue":"3","key":"12209_CR118","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1016\/0304-4076(86)90063-1","volume":"31","author":"T Bollerslev","year":"1986","unstructured":"Bollerslev T (1986) Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics 31(3):307\u2013327","journal-title":"Journal of Econometrics"},{"key":"12209_CR119","unstructured":"Oreshkin BN, Carpov D, Chapados N, Bengio Y (2019) N-beats: Neural basis expansion analysis for interpretable time series forecasting. arXiv preprint arXiv:1905.10437"},{"issue":"1","key":"12209_CR120","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1080\/00031305.2017.1380080","volume":"72","author":"SJ Taylor","year":"2018","unstructured":"Taylor SJ, Letham B (2018) Forecasting at scale. Am Stat 72(1):37\u201345","journal-title":"Am Stat"},{"key":"12209_CR121","unstructured":"Smith TG (2017) pmdarima: Arima estimators for python"},{"key":"12209_CR122","unstructured":"Sheppard K. ARCH: ARCH and Other Tools for Financial Econometrics. https:\/\/github.com\/bashtage\/arch"},{"issue":"8","key":"12209_CR123","doi-asserted-by":"crossref","first-page":"1798","DOI":"10.1109\/TPAMI.2013.50","volume":"35","author":"Y Bengio","year":"2013","unstructured":"Bengio Y, Courville A, Vincent P (2013) Representation learning: a review and new perspectives. IEEE Trans Pattern Anal Mach Intell 35(8):1798\u20131828","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"12209_CR124","doi-asserted-by":"crossref","DOI":"10.1201\/EBK1439826119","volume-title":"Knowledge Discovery from Data Streams","author":"J Gama","year":"2010","unstructured":"Gama J (2010) Knowledge Discovery from Data Streams. Chapman and Hall\/CRC, Boca Raton, FL"}],"container-title":["Neural Computing and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-026-12209-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00521-026-12209-6","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00521-026-12209-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,6]],"date-time":"2026-07-06T05:37:22Z","timestamp":1783316242000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00521-026-12209-6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6]]},"references-count":124,"journal-issue":{"issue":"12","published-print":{"date-parts":[[2026,6]]}},"alternative-id":["12209"],"URL":"https:\/\/doi.org\/10.1007\/s00521-026-12209-6","relation":{},"ISSN":["0941-0643","1433-3058"],"issn-type":[{"value":"0941-0643","type":"print"},{"value":"1433-3058","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6]]},"assertion":[{"value":"23 December 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 May 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 June 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare that they have no conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of Interest"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval"}},{"value":"Not applicable.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent to participate"}},{"value":"Not applicable.","order":5,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The source code for T2f is publicly available at\n                      \n                      .","order":6,"name":"Ethics","group":{"name":"EthicsHeading","label":"Code availability"}}],"article-number":"510"}}