Joshua Dimasaka

I'll be at the 40th #AAAI in Singapore, presenting two research initiatives that advance disaster risk science w/ geospatial artificial intelligence and Earth Observation data! See you!

I'll also be looking for jobs!

Missed our presentation at the 20th #APRU Multi-Hazards Symposium 2025? Here's a short video animation of our work: youtu.be/mwR8xY6fvBo

[AGU 2025] Data-Driven Probabilistic Approach ... Towards Global Disaster Risk Quantification Audit YouTube video by Joshua Dimasaka

And, a video narration too!
📹 youtu.be/Ryu-_We6fMo

Now available online!
🌐 agu25.ipostersessions.com/Default.aspx...

👋 Curious about AI4ER-CDT research on dynamic & probabilistic modelling for spatiotemporal disaster risk assessment? I'll be online at @agu.org 2025 (#AGU25) in New Orleans, Louisiana, & Online World!

🗓️ 14:15-17:45 (CST), Tuesday, Dec 16th
🌐 iPoster: agu25.ipostersessions.com/Default.aspx...

View of the stage, where the awardees are presented.

Oluwaseunnla Precious Adelusi speaks at the ceremony.

Rainer Müssner from the German ministry of science greets the participants of the ceremony.

At the Digital GreenTalents Award Ceremony 2025, 20 young researchers received this year's award for their work at the intersection of #digitalization and #sustainability.

We're looking forward to seeing all the talents at the Summer School 2026!

Learn more: digitalgreentalents.de/awardees-2025

💼 Lastly -- hello again, job market, I’m finally back! Feel free to send me any academic, industry, and/or consulting roles for 2026 (open to work anywhere) at jtd33@cam.ac.uk!

🇵🇭 20th APRU Multi-hazards Symposium & Conference, Philippines
Today, I presented our city-scale work, which is a fruitful collaboration with city disaster risk managers, EMI (a nonprofit scientific organization) in Quezon City, and our university.

🇨🇾 8th Int’l Disaster & Risk Conference 2025, Best Overall Conference Paper, Cyprus
We present a weakly supervised first-order transition matrix that reflects the changes in the spatiotemporal distribution of physical vulnerability in two disaster-stricken and socioeconomically disadvantaged areas.

to support the 2015-2030 UNDRR Sendai Framework! I’ll also be at the job fair, looking for a job (so let’s chat)!

Long hours of careful thinking, previous experience on regional risk modelling, and a deep understanding of the intersection of disaster risk science and AI & EO have guided us to develop a framework that can help us assess regional disaster resilience ...

🇸🇬 AAAI AI for Social Impact 2026, Singapore
As a first-timer to attend the main tech track part of an AI conference, I’m excited that our PhD research has been selected as an "Oral Presentation" for AI for Social Impact (AISI) at the 40th AAAI in Singapore, Jan 22-25, 2026.

🇩🇪 Digital GreenTalents, Germany
Honored to be one of the 20 selected young scientists all over the world to receive the “Digital GreenTalents - High Potentials in Sustainable Development Award 2025” by the German Federal Ministry of Research, Technology, and Space (BMFTR)

Hello! Time for my year-ender updates, as I'm now in my final year as a PhD researcher at the University of Cambridge 🇬🇧 (mentored by Emily So, FICE, Christian Geiß, Robert Muir-Wood, Fouad Bendimerad) in applying AI & EO to the study of environmental risks (AI4ER) in the built environment (CURBE):

📜 DeepC4: Deep Conditional Census-Constrained Clustering for Large-scale Multitask Disaggregation of Urban Morphology
🌐 lnkd.in/gC7Nby2n (arXiv preprint)

📜 A Data-Driven Probabilistic Approach to Regional Dynamics of Building Exposure and Physical Vulnerability Towards Global Disaster Risk Quantification Audit
🏢 AGU25, Progress, Ethics, and Accuracy in Bldg Stock Attribution (NH23D)
📍 New Orleans, LA, US
🗓 Dec 15-19, 2025

📜 Modernizing Quezon City's Building Risk Data with AI and Earth Observation
🏢 20th Assoc. of Pacific Rim Universities Multi-hazards Conference and Symposium (APRU-MH20)
📍 UP BGC, Philippines
🗓 Nov 26-29, 2025

📜 GraphCSVAE: Graph Categorical Structured Variational Autoencoder for Spatiotemporal Auditing of Physical Vulnerability Towards Sustainable Post-Disaster Risk Reduction
🏢 8th Int'l Disaster and Risk Conference (IDRC)
📍 Nicosia, Cyprus
🗓 Oct 22-24, 2025

🌏 On the 𝗜𝗻𝘁𝗲𝗿𝗻𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗗𝗮𝘆 𝗳𝗼𝗿 𝗗𝗶𝘀𝗮𝘀𝘁𝗲𝗿 𝗥𝗶𝘀𝗸 𝗥𝗲𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝟮𝟬𝟮𝟱, we, at the University of Cambridge, have been conducting timely research in support of the 2015-2030 UNDRR #Sendai Framework using #AI and #EarthObservation data. #DRRday 🧵

"In the years following a disaster, do affected communities reduce their future risk by avoiding the construction of vulnerable structures, or do they rather return to settle in pre-identified danger zones, even after experiencing their devastating impacts firsthand?"

using cluster-level constraints, multitask optimization, and multilabel conditionality, under a weak supervision setup.

#HelmholtzAI #AI4ER #DeepClustering #EO #UrbanMorphology #DisasterRisk #SpatialAudit #SpatialDisaggregation #Constraints

Our proposed novel DeepC4, a deep constrained clustering model, advances the existing classical spatial disaggregation techniques using deep learning and EO data to understand the distribution of various urban morphology characteristics at large scales, ...

Our non-peer-reviewed preprint "Deep Conditional Census-Constrained Clustering (DeepC4) for Large-scale Multitask Disaggregation of Urban Morphology" is now available on arXiv! This work is the highlight contribution of my first-year PhD-ing, which deals with "spatial auditing".

✅ 𝗥𝗲𝗴𝗶𝘀𝘁𝗿𝗮𝘁𝗶𝗼𝗻: apply.knight-hennessy.stanford.edu/register/?id...

Mayroon po ba kayong mga tanong sa ating mga iskolar? Ling Ritter and I will be doing our 𝟯𝗿𝗱 𝗢𝗻𝗹𝗶𝗻𝗲 𝗦𝗰𝗵𝗼𝗹𝗮𝗿 𝗤&𝗔 on 𝗝𝘂𝗻𝗲 𝟮𝟵𝘁𝗵, 𝗦𝘂𝗻𝗱𝗮𝘆, 𝟭𝟮-𝟭𝗽𝗺 (𝗣𝗵𝗶𝗹𝗶𝗽𝗽𝗶𝗻𝗲 𝗦𝘁𝗮𝗻𝗱𝗮𝗿𝗱 𝗧𝗶𝗺𝗲), in a mix of English, Filipino/Tagalog 🇵🇭.

Pakibahagi po! Maraming salamat po.

#PinoyScientists #GraduateScholarship

This is tomorrow! We're excited. Don't forget to register if you'd like to meet our friendly AI4ER community and visit Cambridge for a day with our energizing sunshine ☀️ 🚣 and an afternoon filled with AI + environmental risk research! 🌏 🛰️ 🧠 Kitakits!

some of our newest and exciting updates in how we align the use of AI with the practice of catastrophe modeling and disaster risk science!

#AI #DRR #EnvironmentalRisk #Cambridge #AI4ER #CDT #regional #disaster #risk #catastrophe #EO #ML #CAT #graphML #Bayesian

and physical vulnerability, regionally, given the vast amount of Earth Observation (EO) data and advanced machine learning (ML) techniques like spatiotemporal graph deep learning and Bayesian ML? That's what I will share high-level in our AI4ER CDT Showcase, including ...

When I began my journey in disaster risk science, it was fascinating to understand the mathematics behind probabilistic (seismic, tsunami, flood, blast, wildfire, etc.) hazard analysis. What if, in our risk equation, there's also probabilistic distribution of building exposure ...

Our annual AI4ER CDT Showcase is back this 2025! If you'll be free on May 20th, come and join us at the University of Cambridge (UK) and get to know some of the amazing research we do in the intersection of AI and environmental risks.

RSVP: www.eventbrite.co.uk/e/ai-for-the...