Workshop: Artificial Intelligence and the Environment from The Global South

Indonesia Society for Remote Sensing (MAPIN) and IEEE OESTEMS Indonesia Section collaborating with Cornell Atkinson Center for Sustainability, Department of Information Science and Radiant Earth Foundation present a hybrid 2-day workshop on 13-14 July 2022 09:00 WIB (GMT+7) with theme on Artificial Intelligence and the Environment from the South.

Opening Speeches:

• Dr. Agustan (Chair of MAPIN/ISRS)
• Dr. Marina Frederik (President IEEE OESTEMS Indonesia Section),

Speakers:

• Hamed Alemohammad, Ph.D (Radiant Earth Foundation)
  Radiant MLHub: Advancing the Impact of Machine Learning Applications on Earth Observations

Earth observations (EO) provide consistent data over different spatio-temporal scales that can be used for modeling the Earth and its changing environment. Such modeling efforts require characterization of complex and nonlinear processes that could benefit from data-driven techniques – in particular, machine learning (ML). ML techniques present a game-changing opportunity: the ability to identify and address unique, complex, and emerging challenges at local, regional, and global scales quickly and accurately. From poverty alleviation, food security, and climate change adaptation to sustainable resource management and humanitarian response, the combination of EO with ML can help humanity see, understand, and respond to a rapidly changing world.

ML techniques learn from data and any uncertainty or bias in the data propagates to the model and future estimates. To enable successful applications of ML techniques in Earth science, it’s paramount to build an ecosystem for sharing and publishing training data, models and best practices. Such an ecosystem can make these techniques more accessible, bring transparency and trust in the results, and increase their impact to other problems and geographical regions.

In this presentation, I will introduce Radiant MLHub, the first open-access repository for geospatial training data and models. I will demonstrate the role of benchmarks for advancing applications of EO using ML techniques, and the ways reproducible workflows on Radiant MLHub facilitates this. I will also introduce LandCoverNet a global and geographically diverse human-verified land cover classification training dataset. I will finish by demonstrating the ways to integrate datasets and models from Radiant MLHub into real-world applications for environmental mapping.  

• Cindy Kaiying Lin, Ph.D. (Cornell Atkinson Center for Sustainability)
  Computing and AI from the Global South

In this brief talk, I will introduce the vision and goals of the two-day workshop. I ask participants to imagine what computing and AI from the Global South would look like, especially when such techniques and systems have been used to both impoverish and enrich communities around the world. More importantly, the workshop hopes to develop a community of practitioners from postcolonial countries to use AI in ways that enable them to reassert their own voices and shape their own futures. In a recent MIT Tech Review series entitled, “AI Colonialism”, featured authors brought into attention how the impact of AI reinforces patterns of colonial extraction and exploitation of land, resources, and people. They argue that major tech companies in Euro-American regions repeat similar logics, but through the extraction of data for algorithms. This workshop hopes to flip such a logic by fostering a community of practitioners in the Global South to establish a new vision of AI for the people.

This event will also be presenting Last Lecture from Swasetyo Yulianto, M.Sc. (Overseas Fellowship Program Awardee/OFP 1st batch – Netherland, Indonesia).

Registration: https://s.id/cornell_mapin
Zoom Meeting ID: 863 4709 7925
Passcode: smart

For further information please contact:
Rika Hernawati (MAPIN) +62 813-2071-4281
Zilda Dona (MAPIN) +62 821-8231-4003