首页 > AI前沿 > MarathonTS: a benchmark of long synthetic time series from window assembly and long-series generators

MarathonTS: a benchmark of long synthetic time series from window assembly and long-series generators

arXiv机器学习 2026-11-13 21:58 11 阅读 查看原文
用户希望我作为一个专业的新闻排版工程师,将一段纯文本新闻正文排版成语义化的HTML片段。 规则如下: 1. 只输出HTML片段本身,不要Markdown围栏、不要解释。 2. 每个自然段落用`

`包裹,长文用`

`/`

`划分小节。 3. 列表内容用`
  • `,引用用`
    `,代码用`
    `,重要句子可用``。
    4. 保留原文所有信息,不得增删改内容。
    5. 不要输出``/``/``等文档级标签。
    
    原文内容:
    Many applications of synthetic time series require a series that is much longer than the windows on which a generative model is trained. Such a series can be assembled from a pool of short generated windows, which requires deciding which windows are taken, in what order, and how their seams are processed, or it can be produced directly by a long-series generator. This study presents MarathonTS, to our knowledge the first open benchmark that compares both families under one protocol. Every method produces a series of the same length on 9 open hourly datasets with 10 seeds. The series is evaluated by its fidelity to the real data (83 metrics in 12 groups, against three real references: train, validation and an out-of-time split), by its utility (models are trained on the synthetic series only and tested on real data, TSTR against TRTR, in 14 classification and regression tasks), by its memorisation of the training data and by its computational cost. The statistical analysis is carried out as a separate stage that reads the published tables and gives the rankings of every method in every slice, the answers to the research questions with tests and uncertainty, and the results of a pre-specified confirmatory family. In the main run, the methods differ in both confirmatory cells; the choice of windows matters more than the processing of their seams, which does not improve the series, and no method trains the downstream models as well as the real series. The top tier of the leaderboard is reproduced on other hardware, whereas the exact places within it are not. For the reproducibility of the study, the code, the data hashes and all results are publicly available (links below).
    
    分析原文结构:
    这是一段连续的学术/技术新闻/摘要文本,没有明显的小标题或列表。
    可以将其划分为几个逻辑段落,或者加一个小标题。
    由于要求“长文用`

    `/`

    `划分小节”,我可以添加一个`

    `作为大标题,比如 "MarathonTS: 综合时间序列生成基准"(但规则4要求“不得增删改内容”,所以不能随意加标题文字!只能使用原文的内容,或者不加标题,或者把原文的某句话作为标题?不,加标题如果不属于原文就会违反“不得增删改内容”。所以最好不加`

    `,或者只用`

    `包裹每个自然段落。原文其实是一整段。我可以把原文拆分成几个自然段落。 拆分段落: 段落1:Many applications of synthetic time series require a series that is much longer than the windows on which a generative model is trained. Such a series can be assembled from a pool of short generated windows, which requires deciding which windows are taken, in what order, and how their seams are processed, or it can be produced directly by a long-series generator. 段落2:This study presents MarathonTS, to our knowledge the first open benchmark that compares both families under one protocol. Every method produces a series of the same length on 9 open hourly datasets with 10 seeds. 段落3:The series is evaluated by its fidelity to the real data (83 metrics in 12 groups, against three real references: train, validation and an out-of-time split), by its utility (models are trained on the synthetic series only and tested on real data, TSTR against TRTR, in 14 classification and regression tasks), by its memorisation of the training data and by its computational cost. 段落4:The statistical analysis is carried out as a separate stage that reads the published tables and gives the rankings of every method in every slice, the answers to the research questions with tests and uncertainty, and the results of a pre-specified confirmatory family. 段落5:In the main run, the methods differ in both confirmatory cells; the choice of windows matters more than the processing of their seams, which does not improve the series, and no method trains the downstream models as well as the real series. The top tier of the leaderboard is reproduced on other hardware, whereas the exact places within it are not. 段落6:For the reproducibility of the study, the code, the data hashes and all results are publicly available (links below). 检查是否有需要加粗的重要句子: "This study presents MarathonTS, to our knowledge the first open benchmark that compares both families under one protocol." 可以加粗。 "the choice of windows matters more than the processing of their seams, which does not improve the series, and no method trains the downstream models as well as the real series." 可以加粗。 检查是否有列表、引用、代码:没有。 确保没有Markdown围栏,没有解释,没有文档级标签。 排版结果:

    Many applications of synthetic time series require a series that is much longer than the windows on which a generative model is trained. Such a series can be assembled from a pool of short generated windows, which requires deciding which windows are taken, in what order, and how their seams are processed, or it can be produced directly by a long-series generator.

    This study presents MarathonTS, to our knowledge the first open benchmark that compares both families under one protocol. Every method produces a series of the same length on 9 open hourly datasets with 10 seeds.

    The series is evaluated by its fidelity to the real data (83 metrics in 12 groups, against three real references: train, validation and an out-of-time split), by its utility (models are trained on the synthetic series only and tested on real data, TSTR against TRTR, in 14 classification and regression tasks), by its memorisation of the training data and by its computational cost.

    The statistical analysis is carried out as a separate stage that reads the published tables and gives the rankings of every method in every slice, the answers to the research questions with tests and uncertainty, and the results of a pre-specified confirmatory family.

    In the main run, the methods differ in both confirmatory cells; the choice of windows matters more than the processing of their seams, which does not improve the series, and no method trains the downstream models as well as the real series. The top tier of the leaderboard is reproduced on other hardware, whereas the exact places within it are not.

    For the reproducibility of the study, the code, the data hashes and all results are publicly available (links below).

    再次检查规则: 1. 只输出HTML片段本身,不要Markdown围栏、不要解释; -> 满足 2. 每个自然段落用

    包裹,长文用

    /

    划分小节; -> 原文没有明确的小节标题,强行添加会违反“不得增删改内容”的规则,因此只使用

    包裹拆分的自然段落是合理的。 3. 列表内容用

    • ,引用用
      ,代码用
      ,重要句子可用; -> 使用了,其他没有对应内容,不强行添加。
      4. 保留原文所有信息,不得增删改内容; -> 满足,拆分段落不改变文字内容。
      5. 不要输出//等文档级标签。 -> 满足。
      
      确认文本完全匹配,没有拼写错误或遗漏。
      "Many applications of synthetic time series require a series that is much longer than the windows on which a generative model is trained. Such a series can be assembled from a pool of short generated windows, which requires deciding which windows are taken, in what order, and how their seams are processed, or it can be produced directly by a long-series generator." -> 正确
      "This study presents MarathonTS, to our knowledge the first open benchmark that compares both families under one protocol. Every method produces a series of the same length on 9 open hourly datasets with 10 seeds." -> 正确
      "The series is evaluated by its fidelity to the real data (83 metrics in 12 groups, against three real references: train, validation and an out-of-time split), by its utility (models are trained on the synthetic series only and tested on real data, TSTR against TRTR, in 14 classification and regression tasks), by its memori