Computational models and representation for data science

Tensor factorizations. Compressive sampling. Randomized linear algebra. Graph simplifications and multiresolution representations. Transformations and spectral representations. Distributed algorithms.

Learning, modeling, and inference with data

High-dimensional spatio-temporal modeling. Graph signal processing. Theoretical limits. Anomaly detection. Graph learning. Statistical modeling of heterogeneous data types. Post-selection inference. Analysis of deep learning algorithms. Crowdsourcing. Stream mining. Statistical uncertainty quantification.

Acquisition, storage, and retrieval for large-scale data science

Hardware and architectures. Software and Cyberinfrastructure. Protocols for networked storage. Compression for data storage. Sketching and streaming. Scaling up algorithms.

Visualization, summarization, and analytics

Data presentation architectures and dashboards. Data visualization and human perception / cognition. Business intelligence. Data wrangling.

Data science education

Innovative approaches to teaching data science. Data-informed learning theory. Learning analytics.

Data science process and principles

Reproducible research. Open source data science. Workflow. Meta-analysis. Data science ethics. Algorithmic fairness. Bias in science.

Applications

Social media, recommendation systems and collaborative filtering. Defense, intelligence and security. Biology and medicine. Astronomy and other physical sciences. Audio, image, video analytics and computer vision. Urban informatics. Social sciences. Business analytics, forensics and finance. Applications leveraging domain knowledge for data science.