← Back to all news
arXiv Machine Learning By Takumi Fujimoto, Hiroaki Nishi

When Does Online Adaptation Pay on the Edge? A Leakage-Free Evaluation of Warmup, Learning-Rate Selection, and Resource Trade-offs for Time-Series Forecasting

Read the original on arXiv Machine Learning →

The paper investigates when online adaptation benefits edge time‑series forecasting under distribution drift, using a leakage‑free streaming protocol on six public multivariate datasets. It shows that the warmup budget for static baselines and the choice of learning rate can bias perceived adaptation gains, and that a validation‑only procedure selecting warmup and optimizer rates yields Adam outperforming SGD with momentum in most settings. The study also examines accuracy versus adaptation‑state memory and per‑update latency for different adaptation strategies, highlighting parameter‑efficient variants that are nondominated on the memory axis.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Oct 5

AIGS: Adaptive Incremental Gating System for Online Representation Learning in Non-Stationary Data Streams

The paper introduces AIGS, a lightweight Adaptive Incremental Gating System designed for online representation learning in non‑stationary data streams. AIGS uses a Shock Ratio feedback signal to drive a Continuous Plasticity Controller, enabling smooth adaptation between learning plasticity and memory retention while keeping per‑step complexity linear in the feature dimension. Experiments on smart‑city traffic, meteorological, and industrial datasets show that AIGS improves early‑warning lead times, accelerates recovery after abrupt changes, and enhances anomaly recall in noisy environments.

By SiRui He, Kai Liang Lew, Chui Zi Ong, Chean Khim Toa
llms
More like this →
arXiv Machine Learning
Oct 9

AdaptLSTM: Efficient Adaptive Online Learning for Cloud Workload Forecasting under Distribution Drift

AdaptLSTM is an adaptive online learning framework designed for cloud workload forecasting under distribution drift. It detects drift using validation‑calibrated thresholds and performs selective, targeted updates, achieving a large portion of the accuracy gains of naive online learning while reducing per‑step compute cost. Experiments on Alibaba and Container traces show significant efficiency improvements and error reductions compared to static models and classical drift detectors.

By Xinhua Miao, Bowei Yang, Zhengong Cai
llmsefficiency
More like this →
arXiv AI
Sep 16

Schema-Adaptive Action-Conditioned JEPA for Cross-Machine CNC Transfer under Partial Sensor Overlap

The paper introduces a schema‑adaptive action‑conditioned Joint‑Embedding Predictive Architecture (SAAC‑JEPA) for cross‑machine CNC transfer when only a subset of sensors overlap between source and target machines. Experiments show that pretraining does not improve source‑only forecasting, but a carefully selected action‑conditioned JEPA model achieves a zero‑shot RMSE of 0.546 on the target, outperforming persistence but falling short of certain baseline models. Ablation studies reveal that adding RevIN improves RMSE but harms calibration, and limited post‑lock adaptation can further reduce error.

By Ayoub Louaye Bouaziz, Matthieu Ostertag, Anton Demasles
ragbenchmarks
More like this →
arXiv AI
Oct 21

Parameter-Efficient Adaptation of Time Series Foundation Models for Electrical Load Forecasting

The study evaluates how well time‑series foundation models (TSFMs) can be adapted for electrical load forecasting. It compares pretrained inference, full fine‑tuning, and low‑rank adaptation (LoRA) across four load aggregation levels, using MOIRAI 1.0 Large as the main backbone and Chronos‑2 as a secondary model. LoRA achieves a 7.5% average NMAE reduction on MOIRAI while updating only about 1% of the parameters, offering a strong accuracy‑efficiency trade‑off, though benefits vary in out‑of‑distribution scenarios.

By Aliullah, Julian Oelhaf, Edouard Fouch\'e, Jessica Deuschel, Christian Bergler, Andreas Maier, Siming Bayer
fine-tuning
More like this →