Time-series — Topic

📑 arXiv 3d ago

MambaSL: Exploring Single-Layer Mamba for Time Series Classification

MambaSL achieves state-of-the-art time series classification using a single-layer Mamba architecture with TSC-specific modifications. Re-evaluates 20 baselines across all 30 UEA datasets under unified protocol, demonstrating SSMs can excel at time series tasks with minimal architectural complexity.

Models Benchmarks Time-series

📑 arXiv 3d ago

Assessing the Potential of Masked Autoencoder Foundation Models in Predicting Downhole Metrics from Surface Drilling Data

Systematic review of 13 papers finds no existing work applies Masked Autoencoder Foundation Models to predict downhole oil/gas drilling metrics from surface sensor time-series, despite MAEFMs' proven effectiveness in time-series modeling. Current approaches rely on ANNs and LSTMs but struggle with scarce labeled downhole measurements.

Time-series Foundation-models Domain-specific

MambaSL: Exploring Single-Layer Mamba for Time Series Classification ↗

Assessing the Potential of Masked Autoencoder Foundation Models in Predicting Downhole Metrics from Surface Drilling Data ↗

MambaSL: Exploring Single-Layer Mamba for Time Series Classification

Assessing the Potential of Masked Autoencoder Foundation Models in Predicting Downhole Metrics from Surface Drilling Data