GeoAI Challenge

Everything happens somewhere – applying machine learning to geospatial analysis

Join the GeoAI Challenge in 2023 (second edition), a competition aimed at providing solutions for collaboratively addressing real-world geospatial problems by applying artificial intelligence (AI)/machine learning (ML).  Through this platform, participants will attempt to address the UN Sustainable Development Goals (SDGs) related problems using real-world data. In addition, participants will acquire hands-on experience in AI/ML in areas relevant to solving SDGs and compete for prizes, recognition, and certificates.

Join the GeoAI Challenge in 2023 (second edition), a competition aimed at providing solutions for collaboratively addressing real-world geospatial problems by applying artificial intelligence (AI)/machine learning (ML).  Through this platform, participants will attempt to address the UN Sustainable Development Goals (SDGs) related problems using real-world data. In addition, participants will acquire hands-on experience in AI/ML in areas relevant to solving SDGs and compete for prizes, recognition, and certificates.

Compute platform

ITU provides a state-of-the-art, free-of-charge compute platform to participants of the Challenge who do not have adequate access to compute in their respective institutions. The compute platform will provide participants with access to:

 

  • Free GPUs and CPUs
  • Hosted Jupyter notebook server
  • Python kernel
  • Pre-installed machine learning packages, e.g. PyTorch and Tensorflow

 

In some of the problem statements, a baseline or reference solution may be offered which may include implementations using Jupyter notebooks.

The GeoAI Challenge features five problem statements

Landslide Susceptibility Mapping

Develop ML algorithms that can analyze large dataset to identify patterns indicating high probability of landslide occurrence and create a landslide susceptibility map.

Curated by GEOlab at Polytechnic di Milano

Cropland Mapping

Develop accurate, cost-effective classification model for cropland extent mapping with ML techniques in three test regions.

Curated by UNODC (United Nations Office on Drugs and Crime) and FAO (Food and Agriculture Organization of the United Nations)

Air Pollution Susceptibility Mapping

Implement a machine learning method which can accurately estimate the pollution levels (AQI) of the metropolitan city of Milan

Curated by GEOlab at Polytechnic di Milano

The Hyperview Challenge

Estimating soil parameters from hyperspectral images.

Curated by ESA (European Space Agency)

Location Mention Recognition (LMR)

This challenge aims at automatically extracting toponyms (places or location names) from the given text.

Curated by QCRI (Qatar Computing Research Institute), QU (Qatar University), and Qen Labs Inc.

GeoAI Challenge Timeline

February – June 2023

July – October 2023

November – December 2023

Curation Phase

Competition Phase

Evaluation Phase

GeoAI Challenge Timeline

February – June 2023

Curation Phase

July – October 2023

Competition Phase

November – December 2023

Evaluation phase

Everything happens somewhere

Sky

GEOAI-DataSources-SKY-01

Ground

GEOAI-DataSources-GROUND-01
GEOAI-DataSources-GROUND-UserGen-03

Water + below surface level

GEOAI-DataSources-BELOWGROUND-01
Related sessions
15 December 2023
11:00 - 12:00 CET Geneva | 05:00-06:00 EST, New York | 18:00-19:00 CST, Beijing
Jakub Nalepa (Silesian University of Technology), Nicolas Longepe (ESA), Andrea Manara (ITU)
23 February 2024
16:00 - 17:00 CET Geneva | 10:00-11:00 EST, New York | 23:00-00:00 CST, Beijing
Di Zhu (University of Minnesota-Twin Cities), Song Gao (University of Wisconsin-Madison), Wenwen Li (Arizona State University)...
1 March 2024
15:30 - 17:00 CET Geneva | 09:30-11:00 EST, New York | 22:30-00:00 CST, Beijing
Blagoj Delipetrev (European Commission's Joint Research Centre), Maria Antonia Brovelli (Politecnico di Milano)

Benefits

geoai-ben1

Crowdsourcing multiple solutions for high-impact problems that could improve real production AI/ML systems

geoai-ben2

Increasing awareness about the problem domain and your work either in research or business

geoai-ben3

Access to a growing and highly skilled talent pool of AI researchers, students, and professionals interested in the same problems as your group for further collaborations or hiring

Technical partners

AIIA-300x
JLLogoCaption1x1

Sponsorship Inquiries​






    Are you sure you want to remove this speaker?