Feb 26, 2024 · Our method, TOTEM, or TOkenized Time Series EMbeddings, proposes a simple tokenizer architecture that embeds time series data from varying domains.
TOTEM explores time series unification through discrete tokens (not patches!!). Its simple VQVAE backbone learns a self-supervised, discrete, codebook.
Mar 8, 2024 · A simple, performant, time series tokenizer that works across domains and tasks thereby enabling generalist modeling with strong in-domain and zero-shot ...
Our proposed discrete time series tokenization enables the design of general models across a variety of time series domains, tasks, and evaluation schemas, ...
Feb 26, 2024 · We present TOTEM, a simple tokenizer architecture for time series analysis that works across domains and tasks with minimal to no tuning. Report ...
Feb 26, 2024 · TOTEM, or TOkenized Time Series EMbeddings, proposes a simple tokenizer architecture that embeds time series data from varying domains.
Feb 29, 2024 · TOTEM: TOkenized Time Series EMbeddings for General Time Series Analysis TOTEM learns discrete time tokens (not patches!!) Image. 6:16 PM ...
3 days ago · TOTEM: TOkenized Time Series EMbeddings for General Time Series... This work studies the problem of time series analysis with generalist models, ...
Feb 26, 2024 · TOTEM: TOkenized Time Series EMbeddings for General Time Series Analysis ; common architectural backbone can be retrained on a specific task for ...
TOTEM: TOkenized Time Series EMbeddings for General Time Series Analysis. S Talukder, Y Yue, G Gkioxari. arXiv preprint arXiv:2402.16412, 2024. 1, 2024.