Abstract:Local temporal patterns in real-world time series continuously shift, rendering globally shared transformations suboptimal. Current deep forecasting models, despite their scale and complexity, rely on fixed weight matrices applied uniformly to all temporal tokens. This creates a static pattern response: models settle into a compromised average, unable to adapt to changing local dynamics. We introduce Dynamic Pattern Recalibration (DPR), a backbone-agnostic mechanism that resolves this via token-level recalibration. Through a lightweight "Perceive-Route-Modulate" pipeline, DPR computes a soft-routing distribution over a learned basis of adaptive response patterns, generating a time-aware modulation vector that recalibrates hidden states via a residual Hadamard product. As a backbone-agnostic adapter, DPR enhances forecasting across diverse architectures with minimal overhead, confirming it addresses a general bottleneck. As a minimalist standalone model, DPRNet achieves competitive performance across 12 benchmarks, validating dynamic recalibration against macroscopic parameter scaling.
| Comments: | 22 pages, 6 figures. Preprint |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.06310 [cs.LG] |
| (or arXiv:2605.06310v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.06310 arXiv-issued DOI via DataCite (pending registration) |
From: Siru Zhong [view email]
[v1]
Thu, 7 May 2026 14:12:47 UTC (1,090 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。