CITY-AI CyberTraining for AI-Enabled Intelligent Urban Informatics and Smart City Innovation

CITY-AI Education Modules

The CITY-AI curriculum aims to bridge the gap between urban science, cyberinfrastructure (CI), and artificial intelligence (AI) using the HPC resources including I-GUIDE Platform and NSF Anvil Supercomputer.

Curriculum Overview

All training materials are developed as Open Educational Resources (OERs) following FAIR (Findability, Accessibility, Interoperability, and Reusability) principles, packaged with interactive Jupyter Notebooks and containerized Docker environments.

CITY-AI Urban Informatics Curriculum Themes

Theme 1: Core CI Skills

Linux, HPC, Python, Profiling & Scaling, NSF Anvil Supercomputer

Theme 2: Urban Analytics

Spatial Data, Big Data, Streaming IoT, Spatiotemporal Analysis

Theme 3: Urban Use-Cases

Air Quality, Urban Heat Islands, GeoAI, SVI & Computer Vision

Theme 1: Core Cyberinfrastructure (CI) Skills

These modules lower technical barriers for urban planners and geographers by introducing high-performance computing (HPC) environments, terminal workflows, and distributed resources.

Module Topic Level Key Learning Goals
Intro to Linux Beginner Terminal navigation, shell scripting, file management, I/O streams, and cluster environment basics.
Intro to HPC Beginner Accessing NSF ACCESS resources, NSF Anvil supercomputer, OpenOnDemand portal, and batch job scheduling (SLURM).
Python & JupyterLab Intermediate Python programming in JupyterLab; scientific libraries including NumPy, SciPy, Pandas, and Matplotlib.
Profiling & Scaling Advanced Computational complexity analysis, estimating resource requirements, parallelization, and software scaling strategies.

Theme 2: Urban Data Analytics

Focused on building fundamental geospatial, spatiotemporal, and high-velocity streaming data processing skills.

Module Topic Level Key Learning Goals
Spatial Data Beginner Core GIS concepts, spatial reference systems, vector/raster data structures, and interactive thematic mapping.
Big Data Intermediate Handling high-volume urban datasets, memory-efficient processing, spatial indexing, and distributed storage.
Streaming Data Intermediate Acquiring, cleaning, and ingesting real-time urban data streams from sensor networks and IoT gateways.
Spatiotemporal (ST) Data Advanced ST analysis techniques, clustering algorithms, spatiotemporal autocorrelation, and dynamic geovisualization.

Theme 3: Urban Informatics Use-Cases

Hands-on case studies demonstrating how cutting-edge GeoAI and CI solve urgent smart city and environmental challenges.

Module Topic Level Key Learning Goals
Volunteered Geographic Info (VGI) Intermediate Extracting, cleaning, and modeling crowdsourced geospatial data from OpenStreetMap (OSM) for urban infrastructure.
Air Quality Monitoring Intermediate Processing spatio-temporal streams from environmental sensors and predicting localized air pollution exposure.
Urban Heat Island (UHI) Advanced Integrating satellite thermal observations with in-situ urban sensor networks to model microclimate temperature anomalies.
Geospatial AI (GeoAI) Intermediate Formulating machine learning models that explicitly account for spatial heterogeneity and temporal dynamics.
Transportation & Accessibility Advanced Utilizing real-time GTFS feeds to analyze public transit delays and compute scalable spatial accessibility metrics.
Computer Vision (CV) Advanced Applying deep learning vision architectures (CNNs, Vision Transformers) to fine-tune image models on urban datasets.
Street View Imagery (SVI) Advanced Automated object detection and segmentation on street-level panoramas to identify urban physical retrofitting and greenery.
Urban Remote Sensing Advanced High-resolution satellite and LiDAR image processing for urban land use classification and 3D canopy extraction.

Computing Environment

All training modules are hosted and executable directly through: