arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.02592cs.LGcs.AI
收藏

通用时间序列预测Transformer的统一训练

Unified Training of Universal Time Series Forecasting Transformers

Salesforce AI Research(Salesforce AI研究院);School of Computing and Information Systems, Singapore Management University(新加坡管理大学计算与信息系统学院)

Gerald Woo, Chenghao Liu, Akshat Kumar, Caiming Xiong, Silvio Savarese, Doyen Sahoo

AI总结:

本文提出Moirai,一种基于掩码编码器的通用时间序列预测Transformer,通过跨频率学习等改进,在大规模数据集LOTSA上训练,实现零样本预测性能媲美全样本模型。

摘要:

时间序列预测的深度学习传统上采用每数据集一个模型的框架,这限制了其利用大型预训练模型带来的变革性影响的潜力。通用预测的概念源于在大量时间序列数据集上的预训练,设想一个单一的大型时间序列模型能够处理各种下游预测任务。然而,构建这样的模型面临时间序列数据特有的挑战:i) 跨频率学习,ii) 适应多变量时间序列的任意变量数量,以及iii) 解决大规模数据中固有的不同分布特性。为应对这些挑战,我们对传统时间序列Transformer架构提出了新颖的改进,从而提出了基于掩码编码器的通用时间序列预测Transformer(Moirai)。我们在新引入的大规模开放时间序列档案(LOTSA)上训练该模型,该档案包含九个领域的超过270亿个观测值,Moirai作为零样本预测器,与全样本模型相比,取得了具有竞争力或更优的性能。代码、数据和模型权重可在https://github.com/SalesforceAIResearch/uni2ts获取。

英文摘要:

Deep learning for time series forecasting has traditionally operated within a one-model-per-dataset framework, limiting its potential to leverage the game-changing impact of large pre-trained models. The concept of universal forecasting, emerging from pre-training on a vast collection of time series datasets, envisions a single Large Time Series Model capable of addressing diverse downstream forecasting tasks. However, constructing such a model poses unique challenges specific to time series data: i) cross-frequency learning, ii) accommodating an arbitrary number of variates for multivariate time series, and iii) addressing the varying distributional properties inherent in large-scale data. To address these challenges, we present novel enhancements to the conventional time series Transformer architecture, resulting in our proposed Masked Encoder-based Universal Time Series Forecasting Transformer (Moirai). Trained on our newly introduced Large-scale Open Time Series Archive (LOTSA) featuring over 27B observations across nine domains, Moirai achieves competitive or superior performance as a zero-shot forecaster when compared to full-shot models. Code, data, and model weights can be found at https://github.com/SalesforceAIResearch/uni2ts.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑