aiqbench analysis
This research presents a scalable framework for estimating time-varying graphs from sparse spatiotemporal data by enforcing constraints on temporal network evolution. The method leverages convex optimization with specific regularization terms to handle real-world datasets like EEG and temperature readings, outperforming existing approaches in signal processing contexts.
Time-Varying Graph Learning with Constraints on Graph Temporal Variation
arXiv:2001.03346v4 Announce Type: replace-cross Abstract: We propose a novel framework for learning time-varying graphs from spatiotemporal measurements. Given an appropriate prior on the temporal behavior of signals, our proposed method can estimate time-varying graphs from a small number of available measurements. To achieve this, we introduce three regularization terms in convex optimization problems that constrain the sparseness of temporal variations of the time-varying networks. Moreover, a computationally scalable algorithm is introduced to solve the optimization problem efficiently. The experimental results with synthetic and real datasets (point cloud, temperature, and EEG data) demonstrate that our proposed method outperforms state-of-the-art methods.
Abstract:We propose a novel framework for learning time-varying graphs from spatiotemporal measurements. Given an appropriate prior on the temporal behavior of signals, our proposed method can estimate time-varying graphs from a small number of available measurements. To achieve this, we introduce three regularization terms in convex optimization problems that constrain the sparseness of temporal variations of the time-varying networks. Moreover, a computationally scalable algorithm is introduced to solve the optimization problem efficiently. The experimental results with synthetic and real datasets (point cloud, temperature, and EEG data) demonstrate that our proposed method outperforms state-of-the-art methods.
Accepted for publication in IEEE Transactions on Signal Processing. Copyright 2026 IEEE. Personal use of this material is permitted
Signal Processing (eess.SP); Machine Learning (cs.LG)
arXiv:2001.03346 [eess.SP]
(or arXiv:2001.03346v4 [eess.SP] for this version)
https://doi.org/10.48550/arXiv.2001.03346 Focus to learn more arXiv-issued DOI via DataCite
IEEE Transactions on Signal Processing, vol. 74, pp. 3017-3032, 2026
https://doi.org/10.1109/TSP.2026.3712068 Focus to learn more DOI(s) linking to related resources
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.
Full text under the publisher's open licence · 228 words
Related coverage
STL-GCS: A Planner-Controller Framework for Signal Temporal Logic via Graphs of Time-varying Convex Sets
arXiv · Jul 22, 2026
Multi-Agent Planning with Spatio-Temporal and Topological Constraints using STL-GO
arXiv · Aug 3, 2026
A2QTGN: Adaptive Amplitude Quantum-Integrated Temporal Graph Network for Dynamic Link Prediction
arXiv · Jul 28, 2026
Article: Runtime-Agnostic AI Workflows: A Pattern for Production Durability and Fast Eval Iteration
InfoQ · Aug 6, 2026