analiza aiqbench
This study rigorously reproduces WEASEL 2.0, a dictionary-based time series classifier, and identifies an inefficiency in its original hyperparameter rules for long datasets. By introducing an adaptive ensemble-size rule that scales with data length and class count, the authors achieve significant reductions in memory usage and training time without compromising accuracy. This work offers practical optimization strategies for deploying lightweight models on temporal data streams where resource constraints are critical.
Revisiting WEASEL 2.0: Reproduction, Sensitivity, and an Adaptive Ensemble-Size Rule
arXiv:2608.18021v1 Announce Type: new Abstract: WEASEL 2.0 is a dictionary-based time series classifier that combines dilated sliding windows with a randomised hyperparameter ensemble and a fixed-size dense feature representation. Two of its hyperparameter choices, the maximum ensemble size and the maximum window size, are specified by simple thresholding rules whose chosen thresholds are not empirically justified in the original paper. In this work we reproduce WEASEL 2.0 on 114 UCR datasets, achieving a mean accuracy of 0.865 and median of 0.928, closely matching the published values (Wilcoxon signed-rank, p = 0.655). We then test the sensitivity of four design choices: the downstream classifier, the absence of feature weighting, the maximum window-size rule, and the maximum ensemble-size rule. The first three are robust to perturbation. The fourth is over-provisioned for long-series datasets, motivating an adaptive rule that sets the maximum ensemble size from series length and number of classes. Evaluated on fixed-length datasets, the adaptive rule reduces peak fit memory by a median of 37 MB (mean 395 MB) and fit time by a median of 0.4 s (mean 4 s), with a median accuracy change of 0% (mean -0.11%). Memory and time savings concentrate on long-series datasets where the original rule allocates the largest ensemble size.
Abstract:WEASEL 2.0 is a dictionary-based time series classifier that combines dilated sliding windows with a randomised hyperparameter ensemble and a fixed-size dense feature representation. Two of its hyperparameter choices, the maximum ensemble size and the maximum window size, are specified by simple thresholding rules whose chosen thresholds are not empirically justified in the original paper. In this work we reproduce WEASEL 2.0 on 114 UCR datasets, achieving a mean accuracy of 0.865 and median of 0.928, closely matching the published values (Wilcoxon signed-rank, p = 0.655). We then test the sensitivity of four design choices: the downstream classifier, the absence of feature weighting, the maximum window-size rule, and the maximum ensemble-size rule. The first three are robust to perturbation. The fourth is over-provisioned for long-series datasets, motivating an adaptive rule that sets the maximum ensemble size from series length and number of classes. Evaluated on fixed-length datasets, the adaptive rule reduces peak fit memory by a median of 37 MB (mean 395 MB) and fit time by a median of 0.4 s (mean 4 s), with a median accuracy change of 0% (mean -0.11%). Memory and time savings concentrate on long-series datasets where the original rule allocates the largest ensemble size.
24 pages, 7 figures. Accepted at the 11th International Workshop on Advanced Analytics and Learning on Temporal Data (AALTD 2026), held at ECML/PKDD 2026, Naples, Italy
Machine Learning (cs.LG)
arXiv:2608.18021 [cs.LG]
(or arXiv:2608.18021v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2608.18021 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 245 cuvinte
Subiecte conexe
How Many Trees in a Random Forest? A Revisited Approach with Plateau Search and Optuna Integration
arXiv · 3 iun. 2026
Game-Theoretic Disempowerment: Loss of Control to Agencyless AI
LessWrong · 27 sept. 2026
Expenditure Horizon: Measuring Optimization Ability, with an Application to NanoGPT
METR · 21 iul. 2026
Async GRPO with LoRA across HF Jobs: a bucket, a proxy, and no NCCL
Hugging Face · 10 sept. 2026