Temporal Data
Temporal data analysis focuses on understanding and modeling data that changes over time, aiming to extract patterns, make predictions, and gain insights from dynamic systems. Current research emphasizes the development and application of advanced machine learning models, including graph neural networks, transformers, and diffusion models, often incorporating physical constraints or leveraging pre-trained language models for improved accuracy and efficiency in tasks like forecasting and anomaly detection. This field is crucial for diverse applications, from predicting traffic flow and disease outbreaks to analyzing financial markets and understanding climate change, driving advancements in various scientific disciplines and practical domains.
Papers
Causal Representation Learning in Temporal Data via Single-Parent Decoding
Philippe Brouillard, Sébastien Lachapelle, Julia Kaltenborn, Yaniv Gurwicz, Dhanya Sridhar, Alexandre Drouin, Peer Nowack, Jakob Runge, David Rolnick
Gridded Transformer Neural Processes for Large Unstructured Spatio-Temporal Data
Matthew Ashman, Cristiana Diaconu, Eric Langezaal, Adrian Weller, Richard E. Turner