HackerPapers

arXiv:2310.10688cs.CL

A decoder-only foundation model for time-series forecasting

Abhimanyu Das, Weihao Kong, Rajat Sen, Yichen Zhou

Abstract

Motivated by recent advances in large language models for Natural Language Processing (NLP), we design a time-series foundation model for forecasting whose out-of-the-box zero-shot performance on a variety of public datasets comes close to the accuracy of state-of-the-art supervised forecasting models for each individual dataset. Our model is based on pretraining a patched-decoder style attention model on a large time-series corpus, and can work well across different forecasting history lengths, prediction lengths and temporal granularities.

arXiv comments
Journal reference
DOI
Subjects
cs.CLComputation and Language cs.AIArtificial Intelligence cs.LGMachine Learning

On Hacker News

Linked by 3 identifiable commenters in 3 comments across 3 threads, from Mar 2024 to Mar 2026.

Comments linking this paper, per year.

Discussed in comments3 threads

  1. Google's 200M-parameter time-series foundation model with 16k context 1 comment
  2. TimesFM: Time Series Foundation Model for time-series forecasting 1 comment

    • Third paragraph of the introduction of the mentioned paper[1] in the first paragraph of the repo. [1] https://arxiv.org/abs/2310.10688

      melenaboijaView on HNParent context

  3. Moirai: A time series foundation model for universal forecasting 1 comment

    • Interesting, but I'm very skeptical. There are over a dozen transformers-based foundation time series model released in the past year and without fail, every one of them claims to be at or near SOTA. For example: - Time-LLM (https://arxiv.org/abs/2310.01728) - Lag-Llama…

      longdogView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.