Towards Actionable Spatiotemporal Time series Forecasting

الموضوعات

علوم تطبيقية وتكنولوجية

الملخص

Spatiotemporal forecasting systems have become indispensable tools for decision making across domains such as transportation, earth science, and urban planning. Their utility is continuously growing, especially with the growth of computing power and the advent of powerful Deep Learning (DL) based data-driven methods capable of modeling complex systems. However, a significant gap exists between forecasts that optimize for prevalent error metrics and actionable forecasts that enable effective operational decision-making. This dissertation addresses this gap by developing methodologies that transform spatiotemporal forecasting from models optimized for statistical accuracy to those designed for real-world utility. The work establishes a framework for actionable spatiotemporal forecasting, identifying limitations in current DL approaches that rely on pixel-wise error metrics like Mean Squared Error, which encourage temporal averaging that smooths out crucial variations, precisely the features most valuable for decision-making. Two major contributions address actionability: first, a novel loss function incorporating Total Variation regularization improves perceptual quality by preserving important temporal dynamics while maintaining distortion performance across multiple architectures and datasets; second, FiLM-SimVP, a new quantile regression model that efficiently quantifies uncertainty in spatiotemporal forecasts with minimal computational overhead, providing well-calibrated uncertainty estimates that scale to complex datasets. These advancements establish a foundation for forecasting systems that balance computational efficiency, representational power, and uncertainty awareness, enabling a shift toward models that capture meaningful patterns essential for real-world deployments of such systems.

روابط وملفات

التعريف والنوع

رقم الوثيقة

0f48d036-5b5d-41cb-b6a2-947095536ec0

رقم العقد

0

نوع الوسائط

Crawler

نوع المحتوى

الرسائل العلمية

صيغة المصدر

رسائل دكتوراة

نوع الملف

pdf text

أسماء الملفات

0f48d036-5b5d-41cb-b6a2-947095536ec0_1.pdf

بيانات النشر

ألقاب المؤلفين

[{"name_ar":" Abdesselam Bouzer-doum","title_ar":"اشراف","title_en":"Supervision"},{"name_ar":"Shehel Yoosuf","title_ar":"اعداد","title_en":"Preparation"}]

اللغة

English

المصدر والدورية

اسم المصدر

Towards Actionable Spatiotemporal Time series Forecasting

المحتوى والصفحات

عدد الصفحات

0

كلمات الباحثين

Computer vision - Machine learning - Spatiotemporal prediction - Timeseries forecasting - Uncertainty quantification

إشراف وإعداد

الإشراف

Abdesselam Bouzer-doum

الإعداد

Shehel Yoosuf

الاقتباسات الببليوغرافية