analiza aiqbench
This research introduces a hybrid Convolutional-Transformer architecture specifically designed to handle the complex spatio-temporal dynamics of Antarctic sea ice forecasting. By integrating month-aware positional encodings and seasonal temporal biases, the model effectively captures long-range dependencies that traditional convolutional or recurrent networks miss. The study validates that combining local feature extraction with global attention mechanisms, augmented by periodic prior knowledge, yields consistent performance gains across short- and long-horizon predictions.
Season-Aware Hybrid Convolutional-Transformer for Antarctic Sea Ice Concentration Forecasting
arXiv:2608.30654v1 Announce Type: new Abstract: Antarctic sea ice concentration (SIC) forecasting is an important yet challenging task due to the coexistence of complex spatial structure, long-range temporal dependencies, and strong seasonal variability. Conventional convolution-based models are effective at capturing local spatial patterns, but often have limited ability to model long-term temporal evolution. To address these challenges, we build on a hybrid Convolutional-Transformer forecasting framework for monthly Antarctic SIC forecasting. This framework combines convolutional encoding for spatial feature extraction with factorised self-attention for spatio-temporal dependency modelling. We further introduce two seasonal prior mechanisms: a month-aware positional encoding that injects calendar-month information into the token representation, and a seasonal temporal bias that encourages attention to periodically related historical states. Experimental results show that the proposed framework achieves better performance than convolutional and recurrent baselines across both classification and regression metrics. Ablation studies further indicate that the seasonal prior mechanisms provide consistent additional gains in both short- and long-horizon prediction. These results demonstrate the value of combining convolutional structures, attention mechanisms, and periodic prior information for Antarctic SIC forecasting.
Abstract:Antarctic sea ice concentration (SIC) forecasting is an important yet challenging task due to the coexistence of complex spatial structure, long-range temporal dependencies, and strong seasonal variability. Conventional convolution-based models are effective at capturing local spatial patterns, but often have limited ability to model long-term temporal evolution. To address these challenges, we build on a hybrid Convolutional-Transformer forecasting framework for monthly Antarctic SIC forecasting. This framework combines convolutional encoding for spatial feature extraction with factorised self-attention for spatio-temporal dependency modelling. We further introduce two seasonal prior mechanisms: a month-aware positional encoding that injects calendar-month information into the token representation, and a seasonal temporal bias that encourages attention to periodically related historical states. Experimental results show that the proposed framework achieves better performance than convolutional and recurrent baselines across both classification and regression metrics. Ablation studies further indicate that the seasonal prior mechanisms provide consistent additional gains in both short- and long-horizon prediction. These results demonstrate the value of combining convolutional structures, attention mechanisms, and periodic prior information for Antarctic SIC forecasting.
Machine Learning (cs.LG)
arXiv:2608.30654 [cs.LG]
(or arXiv:2608.30654v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2608.30654 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 265 cuvinte
Subiecte conexe
RhyMix: A Lightweight Adaptive Multi-Rhythm Network for Long-Term Time Series Forecasting
arXiv · 15 oct. 2026
Links #5: 2026/07
LessWrong · 31 iul. 2026
GraphSVR: A Graph Convolutional Support Vector Regression Framework for Robust Spatiotemporal Air Pollution Forecasting
arXiv · 25 aug. 2026
HyBDM: Multi-Scale Hybrid Experts for Time Series Forecasting with Bidirectional Dependency Modeling
arXiv · 21 iul. 2026