Workshop
6th Workshop on Artificial Intelligence for Humanitarian Assistance and Disaster Response
Ritwik Gupta · Thomas Manzini · Robin Murphy · Eric Heim · Bertrand Saux · Katie Picchione
Room 240 - 241
Natural disasters are one of the oldest threats to both individuals and the societies they co-exist in. As a result, humanity has ceaselessly sought way to provide assistance to people in need after disasters have struck. Further, natural disasters are but a single, extreme example of the many possible humanitarian crises. Disease outbreak, famine, and oppression against disadvantaged groups can pose even greater dangers to people that have less obvious solutions. In this proposed workshop, we seek to bring together the Artificial Intelligence (AI) and Humanitarian Assistance and Disaster Response (HADR) communities in order to bring AI to bear on real-world humanitarian crises. Through this workshop, we intend to establish meaningful dialogue between the communities.By the end of the workshop, the NeurIPS research community can come to understand the practical challenges of aiding those who are experiencing crises, while the HADR community can understand the landscape that is the state of art and practice in AI. Through this, we seek to begin establishing a pipeline of transitioning the research created by the NeurIPS community to real-world humanitarian issues.
Schedule
Fri 7:00 a.m. - 7:15 a.m.
|
Opening Remarks
(
Speaker
)
>
SlidesLive Video |
🔗 |
Fri 7:15 a.m. - 7:45 a.m.
|
Leila Hashemi Beni - North Carolina A&T
(
Invited Speaker
)
>
SlidesLive Video |
🔗 |
Fri 7:45 a.m. - 8:00 a.m.
|
FloodBrain: Flood Disaster Reporting by Web-based Retrieval Augmented Generation with an LLM
(
Oral
)
>
SlidesLive Video Fast disaster impact reporting is crucial in planning humanitarian assistance. Large Language Models (LLMs) are well known for their ability to write coherent text and fulfil a variety of tasks relevant to impact reporting, such as question answering or text summarization. However, LLMs are constrained by the knowledge within their training data and are prone to generating inaccurate, or "hallucinated”, information. To address this, we introduce a sophisticated pipeline embodied in our tool FloodBrain, specialized in generating flood disaster impact reports by extracting and curating information from the web. Our pipeline assimilates information from web search results to produce detailed and accurate reports on flood events. We test different LLMs as backbones in our tool and compare their generated reports to human-written reports on different metrics. Similar to other studies, we find a notable correlation between the scores assigned by GPT-4 and the scores given by human evaluators when comparing our generated reports to human-authored ones. Additionally, we conduct an ablation study to test our single pipeline components and their relevancy for the final reports. With our tool, we aim to advance the use of LLMs for disaster impact reporting and reduce the time for coordination of humanitarian efforts in the wake of flood disasters. |
Grace Colverd · Leonard Silverberg · Paul Darm · Noah Kasmanoff 🔗 |
Fri 8:00 a.m. - 8:30 a.m.
|
Morning Break
(
Break
)
>
|
🔗 |
Fri 8:30 a.m. - 9:00 a.m.
|
Jack Langerman - More Structured Structure from Motion - Semantically Meaningful Reconstruction from Ground Level Images
(
Invited Speaker
)
>
SlidesLive Video |
🔗 |
Fri 9:15 a.m. - 9:30 a.m.
|
A Bayesian Probabilistic Model for Estimating Building Damage Distribution from Forecasts
(
Oral
)
>
SlidesLive Video Estimating the distribution of building damage from forecasted hazards can enable more effective urban search and rescue (USAR) planning and operations. Our work uses Bayesian probabilistic modeling to estimate the distribution of building damage levels from forecasted hazard scores. We leverage data on building damage collected from FEMA USAR teams during wide area search operations. Our results validate the performance of the FEMA hazard forecasts predictions for Hurricane Ian in 2022, and also quantitatively characterizes the distribution of building damage for each forecasted hazard level. |
Jeffrey Liu · Chad Council 🔗 |
Fri 9:15 a.m. - 9:30 a.m.
|
Leveraging AI for Natural Disaster Management : Takeaways From The Moroccan Earthquake
(
Oral
)
>
SlidesLive Video The devastating 6.8-magnitude earthquake in Al Haouz, Morocco in 2023 prompted critical reflections on global disaster management strategies, resulting in a post-disaster hackathon, using artificial intelligence (AI) to improve disaster preparedness, response, and recovery. This paper provides (i) a comprehensive literature review, (ii) an overview of winning projects, (iii) key insights and challenges, namely real-time open-source data, data scarcity, and interdisciplinary collaboration barriers, and (iv) a community-call for further action. |
Léna Néhale Ezzine · Yoshua Bengio · Ayoub Atanane · Ghait Boukachab · Oussama Boussif · · Yassine Yaakoubi · Loubna Benabbou · Léonard Boussioux · Peetak Mitra · Alexandre Jacquillat · Dick Den Hertog · Mehdi Bennis · Ilham EL BOULOUMI · Ayoub Loudyi · Aymane El Firdoussi · Achraf Sbai · SANAE ATTAK · Kaoutar Lakdim · Yassine Squalli Houssaini · Firdawse Guerbouzi · Chaimae Biyaye · Khadija Bayoud · Ikram Belmadani · Charles Bricout · Reyad OUAHI · Alex Maggioni · B.V. Alaka · Kiruthika Subramani · Tariq Daouda · Redouane Lguensat · Khaoula Chehbouni · Afaf Taik · Kanishk Jain · Hamza Ghernati · Lamia Salhi · Laila Salhi · Jules Lambert · Jeremy Pinto · Victor Schmidt · Zhor KHADIR · Nouamane Tazi · Yuyan Chen · Nikhil Reddy Pottanigari · Santhoshi Ravichandran · Ashwini Rajaram · Alex Hernandez-Garcia · Reda Snaiki · Laurent Barcelo · Salim Chemlal · Omar El Housni · AJ Dhimine · Abderrahim Khalifa ·
|
Fri 9:30 a.m. - 10:10 a.m.
|
RescueNet Challenge Summary and Q&A
(
Challenge Summary
)
>
SlidesLive Video |
🔗 |
Fri 10:10 a.m. - 11:30 a.m.
|
Lunch Break
(
Break
)
>
|
🔗 |
Fri 11:30 a.m. - 12:00 p.m.
|
Anna Sobiewska - FEMA Recovery GIS, Recovery Reporting and Analytics Division
(
Invited Speaker
)
>
SlidesLive Video |
🔗 |
Fri 12:00 p.m. - 12:15 p.m.
|
Forecasting Post-Wildfire Vegetation Recovery in California using a Convolutional Long Short-Term Memory Tensor Regression Network
(
Oral
)
>
SlidesLive Video The study of post-wildfire plant regrowth is essential for developing successful ecosystem recovery strategies. Prior research mainly examines key ecological and biogeographical factors influencing post-fire succession. This research proposes a novel approach for predicting and analyzing post-fire plant recovery. We develop a Convolutional Long Short-Term Memory Tensor Regression (ConvLSTMTR) network that predicts future Normalized Difference Vegetation Index (NDVI) based on short-term plant growth data after fire containment. The model is trained and tested on 104 major California wildfires occurring between 2013 and 2020, each with burn areas exceeding 3000 acres. The integration of ConvLSTM with tensor regression enables the calculation of an overall logistic growth rate k using predicted NDVI. Overall, our k-value predictions demonstrate impressive performance, with 50% of predictions exhibiting an absolute error of 0.12 or less, and 75% having an error of 0.24 or less. Finally, we employ Uniform Manifold Approximation and Projection (UMAP) and KNN clustering to identify recovery trends, offering insights into regions with varying rates of recovery. This study pioneers the combined use of tensor regression and ConvLSTM, and introduces the application of UMAP for clustering similar wildfires. This advances predictive ecological modeling and could inform future post-fire vegetation management strategies. |
Jack Liu · Xiaodi Wang 🔗 |
Fri 12:15 p.m. - 12:45 p.m.
|
Kevin R. Nida - FirstNet
(
Invited Speaker
)
>
SlidesLive Video |
🔗 |
Fri 12:45 p.m. - 1:00 p.m.
|
Enabling Decision-Support Systems through Automated Cell Tower Detection
(
Oral
)
>
SlidesLive Video Cell phone coverage and high-speed service gaps persist in rural areas in sub-Saharan Africa, impacting public access to mobile-based financial, educational, and humanitarian services. Improving maps of telecommunications infrastructure can help inform strategies to eliminate gaps in mobile coverage. Deep neural networks, paired with remote sensing images, can be used for object detection of cell towers and eliminate the need for inefficient and burdensome manual mapping to find objects over large geographic regions. In this study, we demonstrate a partially automated workflow to train an object detection model to locate cell towers using OpenStreetMap (OSM) features and high-resolution Maxar imagery. For model fine-tuning and evaluation, we curated a diverse dataset of over 6,000 unique images of cell towers in 26 countries in eastern, southern, and central Africa using automatically generated annotations from OSM points. Our model achieves an average precision at 50% Intersection over Union (IoU) (AP@50) of 81.19 with good performance across different geographies and out-of-sample testing. Accurate localization of cell towers can yield more accurate cell coverage maps, in turn enabling improved delivery of digital services for decision-support applications. |
Natasha Krell · Scott Gleave · Daniel Nakada · justin Downes · Amanda Willet · Matthew Baran 🔗 |
Fri 1:00 p.m. - 1:30 p.m.
|
Afternoon Break
(
Break
)
>
|
🔗 |
Fri 1:30 p.m. - 2:15 p.m.
|
Discussion Panel
(
Discussion Panel
)
>
SlidesLive Video |
🔗 |
Fri 2:15 p.m. - 2:30 p.m.
|
Closing Remarks / Wrap-Up
(
Speaker
)
>
SlidesLive Video |
🔗 |