I'm on the job market for academic & industry research positions! I develop geospatial ML techniques for applications that help tackle climate change and improve our cities. 🤖🌍🌆
Learn more about my research & how to get in touch on my website:
New preprint out on geographic location encoding using spherical harmonics and sinusoidal representations! 🌐➡️🌏
Work led by
@MarcCoru
and w/
@rolf_comma_e
, Robin Zbinden and
@devistuia
.
📄Paper:
💻Code:
🧵Thread: 👇
1/11
Super happy to release 🛰️SatCLIP - a first of its kind global, general purpose location encoder based on Sentinel-2 imagery.
Joint work with
@rolf_comma_e
,
@calebrob6
,
@LesterMackey
&
@MarcCoru
.
📰Paper:
💻Code:
🧵Thread:👇
1/16
Holy crap,
#GameofThrones
is completely falling apart. Some of the worst writing I have ever seen in any TV show. It is heartbreaking seeing the world GRRM built being treated like that. What a shitshow...
Today I passed my PhD defence with minor corrections!
I am thankful to my examiners, Nathan Griffiths and
@jmichaelbatty
for the engaging discussion and to my awesome advisors, Stephen Jarvis, Hongkai Wen and Daniel Neill.
Off to celebrate! 🎊🤩
I'm moving to Boston! 🎉🚀
I am incredibly excited to start as a postdoctoral researcher at
@MSFTResearch
in the New England (
@MSRNE
) lab in fall. I'll continue to work on geospatial problems, improving our cities & tackling climate change with machine learning.🤖🛰️🌏
We're happy to announce the 4th "
#MachineLearning
for the Developing World" workshop at
#NeurIPS2020
. This years edition is focused on building resilience.
We invite contributions from the
#ML4D
community, the submission deadline is Oct 2nd. More info:
I am on the job market for summer / fall 2022, looking for academic (postdoc & faculty) & industry research positions focusing on
#ML
for
#geospatial
data, sustainability &
#urban
analytics.🛰️🌆🌏🤖
You can find my CV & more info on my website:
A new
@Nature
study uses satellite imagery 🛰️ and deep learning 🤖 to map human activity at sea at an unprecedented scale.
In a
@NatureNV
article,
@rolf_comma_e
and I comment on the implications for ML, remote sensing and Earth observation. Check it out below👇
Maps of human activities at sea provide the most complete quantification of ocean industries ever – and reveal a lot of illicit goings-on.
@rolf_comma_e
&
@kklmmr
unpack the findings here
I'll be at
#NeurIPS2022
for the whole coming week! Who's around and up for a coffee / gumbo? Reach out! 🦐
I'll also be at the
@MSFTResearch
booth for a Q&A on my MSR postdoc experience Monday at 1pm, so if you are interested in this stop by:
So excited to see our workshop series growing. Get ready for 133 (yes, no typo) new papers, proposals and tutorials to be presented at the
#NeurIPS2023
@ClimateChangeAI
workshop. Climate-related work is more relevant than ever within the ML community. 🎉🚀🌱
After a record-setting number of submissions, we are excited to announce 130❗️accepted papers & proposals and 3 accepted tutorials at our
#NeurIPS2023
workshop. We cannot wait to see you all in New Orleans!
More info about the workshop here:
The whole
@ClimateChangeAI
workshop at
#NeurIPS2023
will be livestreamed - for free!!!
This is a great opportunity to learn about our growing academic field and to meet and engage with fellow researchers and practitioners. Come join us! 🌱
Are you not able to make it to
#NeurIPS2023
? We're incredibly excited to announce that we will livestream our
@ClimateChangeAI
workshop - for FREE, for EVERYONE. 📺🤓
You can find the livestream on the day of the event (Dec 16) here:
We're happy to (finally😬) release code and slides for SpaceGAN, your friendly neighborhood-conditioned GAN for geospatial point data! Joint work with
@askoshiyama
@flennerhag
:
▶️Code:
▶️Slides:
▶️Paper:
New
@turinginst
colab on spatial data augmentation using spatially-conditioned GANs! We show how GANs can capture both the input and its spatial dependencies.
Joint work with Adriano Koshiyama and
@flennerhag
.
@TensorFlow
implementation coming soon!
We're going to present 🛰️SatCLIP at two
#NeurIPS2023
workshops next week:
▶️Dec 15, DistShift workshop, Room R06-R09 (level 2):
▶️Dec 16,
@ClimateChangeAI
workshop, Great Hall (level 1):
Come by and say hi! 👋
Super happy to release 🛰️SatCLIP - a first of its kind global, general purpose location encoder based on Sentinel-2 imagery.
Joint work with
@rolf_comma_e
,
@calebrob6
,
@LesterMackey
&
@MarcCoru
.
📰Paper:
💻Code:
🧵Thread:👇
1/16
New paper out on improving graph neural networks (GNNs) for geospatial coordinates! Joint work with Daniel Neill and Nathan Safir.🤓🌍
#GeoAI
📰Paper:
💻Code:
🧵Thread: 👇
1/10
Are you at
#AISTATS2023
? Come by my poster today between 4.30-7pm CET! I'm presenting our work on GNNs for geographic data.
#GeoAI
Paper:
Code:
Poster:
🧵:
New paper out on improving graph neural networks (GNNs) for geospatial coordinates! Joint work with Daniel Neill and Nathan Safir.🤓🌍
#GeoAI
📰Paper:
💻Code:
🧵Thread: 👇
1/10
"UN report calls OpenStreetMap “foundational” to disaster risk reduction"
Great to see
@openstreetmap
and
@hotosm
get the attention they deserve for their amazing work.
If you're a PhD student interested in interning with me or one of my amazing colleagues at Microsoft Research New England (
@MSRNE
,
@MSFTResearch
) this summer, please apply here
After a brilliant 6 months at
@ai4eo_de
, I'm back in the UK for the final stages of my PhD. I had a great time with
@xiaoxiang_zhu
and her team and am looking forward to further collaborations during and beyond my PhD 🛰️🌏🤖
Very happy to join the incredible team of
@ClimateChangeAI
, where I hope to combine my passion for
#GeoAI
with the fight against the most pressing issue of our time:
#ClimateChange
. 🌏
The CCAI team is expanding! We are happy to welcome
@realSharonZhou
,
@dwddao
and
@kklmmr
aboard.
With their help, we hope to continue to foster a growing community at the intersection of
#MachineLearning
and
#ClimateChange
.
Stay tuned for upcoming openings within our core team!
As someone digging deeper into Gaussian Processes, I found
@andrewgwils
's collection of resources and tutorials incredibly helpful. I highly recommend checking it our here:
Super exciting new
@turinginst
paper by
@urban_stevie
and
@MateoNeira
on unsupervised feature learning in geospatial / urban settings! It is great to see so much new geography related machine learning work atm.
Very happy to present my new
#GeoML
🌏🛰️ paper: SXL, a framework for better learning with geographic data using spatially explicit auxiliary tasks
Joint work with Daniel Neill at
@NYU_CUSP
.
Paper:
Code:
Thread: 🔽
1/9
My co-author Fernando wrote a blog post based on our paper on information-theoretic community detection for
#bikesharing
. Check it out on
@TDataScience
!
You can find the code including the data to play around with on my Github:
New paper out on GANs for synthetic geospatial coordinates with privacy guarantees.
Joint work with
@tedward78
and
@WarwickDCS
colleagues.
Paper📰
Code💻
Thread🧵👇
1/10
So much work went into this!! We are beyond excited to announce the second edition of the
@ClimateChangeAI
Innovation Grants for impactful ML 🤝 climate projects. Thanks to QCF and
@DeepMind
for supporting this year's grants.🙏
We're excited to announce the 2023
@ClimateChangeAI
Innovation Grants, supported by the Quadrature Climate Foundation &
@DeepMind
.👩🔬💵
We’ll fund projects at the intersection of
#ClimateChange
&
#ML
w/ up to $150k each, for a total of $1.2M.
Learn more:
New (short) paper out today, accepted at
#AIMOCC
workshop @
#ICLR2021
! Preliminary results from our work-in-progress on conditional GANs for simulating extreme weather patterns.
Joint work with
@sudipansaha
, Matthias Kahl,
@linylinx
and
@xiaoxiang_zhu
.
Back in New York! This time for good... (or, well, one year). Excited to spend my third PhD year at
@NYU_CUSP
and the new "Machine Learning for Good" (
#ML4G
) lab 🎉🔬
We are hiring (payed position, not volunteer 💸💸) an events coordinator to lead the
@ClimateChangeAI
summer school! Work on a first of it's kind educational program at the intersection of
#AI
and
#ClimateChange
with us. More info 👇
Passionate about education in
#ClimateChange
and
#AI
? We’re hiring a part-time Educational Events Coordinator to lead & project manage the second edition of our flagship educational event, the CCAI
#SummerSchool
. Work is remote within the US.👩🏫
More info:
Are you at
#NeurIPS2023
and interested in postdoc positions with
@MSFTResearch
? Join me on Dec 14 at 9.15am at the
@Microsoft
booth for a morning coffee ☕️and a postdoc Q&A session.
Full booth schedule is here:
Auxiliary-learning of Moran's I / spatial autocorrelation can help improve neural networks for
#geospatial
data! Paper out now in
#SIGSPATIAL2021
proceedings:
📑(short)
📑(long)
🧵
📺
Very happy to present my new
#GeoML
🌏🛰️ paper: SXL, a framework for better learning with geographic data using spatially explicit auxiliary tasks
Joint work with Daniel Neill at
@NYU_CUSP
.
Paper:
Code:
Thread: 🔽
1/9
Agree 100%. ML research has to acknowledge social impact every step of the way. Taking responsibility shows maturity and a more holistic understanding of the broader real-world impacts. Denying it is lazy, cowardly and ultimately unethical.
@NeurIPSConf
By doing so, more researchers will take part in a broader dialog about issues that reach a much wider audience.
Many simply want to stay in their confined space and let other “experts” deal with ethical/political issues. We all know how that will turn out
New paper out on improving graph neural networks (GNNs) for geospatial coordinates! Joint work with Daniel Neill and Nathan Safir.🤓🌍
#GeoAI
📰Paper:
💻Code:
🧵Thread: 👇
1/10
Excited to be co-organizing the next
@ClimateChangeAI
workshop at
#ICLR2023
! This will also mark the return of an in-person component in our workshop series (though we will maintain a virtual component too).
Can't wait to see you in Kigali!🎉🌍
More info:
We are delighted to invite submissions to our hybrid
#ICLR2023
workshop "Tackling Climate Change with Machine Learning". Important dates:
▶️Mentorship program: Jan 10
▶️Tutorials submissions: Jan 10
▶️Papers / proposals submissions: Feb 5
Learn more:
I wrote a blog post for
@ClimateChangeAI
on our
#AAAI2022
paper on improving GANs with spatio-temporal autoregressive embeddings, joint work w/
@linylinx
et al:
Paper:
Code:
Thread:
New preprint out today on generative modeling of spatio-temporal patterns! 🎉🤖🌎
Jointly lead by
@linylinx
and me, with Beatrice Acciaio (
@ETH_en
) and Daniel Neill (
@nyuniversity
) .
Paper:
Code:
Thread: 👇
(1/11)
I am giving a talk reflecting on my time at
@ai4eo_de
and the broader scope of my PhD research on machine learning for geographic data on July 2nd! 🌏🤖
Come and join 👇
We're excited to announce our next online talk on 02.07.2021 at 09 a.m. with
@xiaoxiang_zhu
and our speaker Konstantin Klemmer
@kklmmr
.Register to this event by writing an email to ai4eo
@tum
.de, or to ai4eo-seminars-subscribe
@lists
.lrz.de for subscribing for info of all lab talks
I cannot recommend this position enough,
@priyald17
's lab is going to be world-leading for
#AI
+
#ClimateChange
research without a doubt. Fantastic opportunity!
I’m recruiting a postdoc to join my (new) group at MIT in Fall 2023!
My group's research will focus on topics at the intersection of machine learning, optimization, control, and power systems, motivated by addressing critical problems of relevance to climate action.
NeurIPS is over but our workshop series continues: We're excited to announce the next edition of "Tackling Climate Change with Machine Learning" at
#ICLR2024
.
Important deadlines:
▶️Mentorship: Jan 3
▶️Tutorials: Jan 10
▶️Papers: Feb 3
More info:
New paper out in
@RSocPublishing
Open Science, with
@NYU_CUSP
,
@unibirmingham
colleagues: "Understanding spatial patterns in rape reporting delays"
This one is really important to me. Check the thread 👇 for a summary.
cw: rape, sexual violence
[1/13]
🌍 Lightning Spotlight on
@Microsoft
's TorchGeo; A PyTorch domain library, similar to torchvision, that provides datasets, transforms, samplers, and pre-trained models specific to geospatial data.
🌟 GitHub:
Super excited about this new
@ClimateChangeAI
initiative: We have set up a volunteer matching platform to help connect NGOs and academic labs with practitioners who want to volunteer their time to climate-related ML projects. Learn more 👇👇👇
Do you need help with
#ML
on a
#climate
-related project? We're setting up a platform to match NGOs and academics with volunteers who are eager to use their skills to tackle climate change. Go to to learn more and submit your project proposal today! 🤓📉👩💻
The Algo2020 Conference - (Nov. 12, 2020)
@ucl
(Online)
Registration Open!
Multi-stakeholder conference on
#AI
& Disruptive Technologies. Check out the variety of session, groups @ world renowned speakers in the line up!
@ai_ucl
@uclcs
@UCLEnterprise
I feel like I am posting a lot of
@ClimateChangeAI
stuff recently, but there is just SO MUCH going on!!!
Next up, our summer school, aimed at
#Climate
folks wanting to learn about
#ML
/
#AI
- and the other way around. Apply by Dec 17! 👇
We are excited to announce our first
@ClimateChangeAI
summer school!🤓👩🎓
Are you an
#AI
expert & want to learn to apply your skills to tackle
#ClimateChange
? Are you a
#Climate
expert & want to learn what tools
#ML
offers?
Apply by Dec 17 & learn more:
This year again, we are organizing AI&Cities track at
@appliedmldays
, one of the biggest and most influential conference on
#AI
. The track will take place on December 2, and call for contributors will soon be opened!
@kklmmr
@EK_geog
#NickJones
#cities
—
I'm going to present our paper on improving GNNs for geographic data with positional encodings at the
@ClimateChangeAI
workshop at
#NeurIPS2022
🎉
Learn more 👇👇👇
New paper out on improving graph neural networks (GNNs) for geospatial coordinates! Joint work with Daniel Neill and Nathan Safir.🤓🌍
#GeoAI
📰Paper:
💻Code:
🧵Thread: 👇
1/10
1/? Beyond our awesome contributed talks, we'll have a bunch of amazing invited talks and further collaborations at
#AIandCities
. In the coming days, I will present some of our highlights on this Twitter thread, so stay tuned:
@appliedmldays
#AMLD2020
Dist Shift workshop talk by
@Reichstein_BGC
was 💯 my highlight of
#NeurIPS2022
. Spatiotemporal dynamics are everywhere & climate change itself is one giant, global dist shift. The ML community has the tools but needs domain expertise from Earth science to address this.
APPLICATIONS NOW OPEN: We're pleased to announce we have opened our call for 2019 doctoral studentships. We're looking for candidates who will embrace the ethos of the Turing and contribute to our vibrant research community. Find out more:
#TuringPhD
The Accepted Papers for
#ML4D2020
are out! We selected 23 submissions from around the world, tackling a range of topics from
#ClimateChange
induced
#Migration
to mitigating
#HIV
transmissions of sex workers.
Check out the papers here:
@ML4Dworkshop
We are very happy to present our
@turinginst
collaboration on GANs for spatial data at the
@UZH_en
workshop on
#ML
for Social & Environmental problems in Zurich this October!
@flennerhag
@askoshiyama
▶️Paper:
▶️Registration: see below
ML4SEP, a peer mentoring group sponsored by
@UZH_en
Graduate Campus, will hold its fall workshop on "Applied ML for Social & Environmental Problems" with a tutorial on SpaceGAN & other sessions on Oct. 3-4, 2019 at Univ. of Zurich, Irchel. Registration:
The recording of my latest talk at
@ai4eo_de
on geospatial machine learning & applications to earth observation and climate is now available on YouTube! Check it out here:
📺
#GeoAI
#MachineLearning
We are excited to organize a
@RealAAAI
Fall Symposium on the role of
#AI
in responding to
#Climate
challenges.
Submissions are due on July 29. Learn more about the program and read the call for submissions here:
New preprint out today on generative modeling of spatio-temporal patterns! 🎉🤖🌎
Jointly lead by
@linylinx
and me, with Beatrice Acciaio (
@ETH_en
) and Daniel Neill (
@nyuniversity
) .
Paper:
Code:
Thread: 👇
(1/11)
The
@ClimateChangeAI
workshop at
@iclr_conf
has evolved into a remote mini-conference with a full week of programme! Check out the amazing speakers and focus areas here:
Very happy to announce our first
#BeyondFellow
Konstantin Klemmer
@kklmmr
, who will stay with us for half year starting from November 1st, 2020.
Interested in visiting us? Apply for your own fellowship here:
We're happy to announce the 4th "
#MachineLearning
for the Developing World" workshop at
#NeurIPS2020
. This years edition is focused on building resilience.
We invite contributions from the
#ML4D
community, the submission deadline is Oct 2nd. More info:
I had a fantastic time at this workshop, learning from an incredibly engaging and driven group of researchers and practitioners. Thanks to
@lilyxu0
,
@rolf_comma_e
and
@MilindTambe_AI
for organizing and structuring the event in a way that felt incredibly productive. 🙏💪
We just had an absolutely awesome workshop on "AI assisted decision making for conservation" at
@Harvard
at
@HCRCS
; will publish our workshop report soon. Big thx to all the amazing participants!! Thx to my co-organizers
@lilyxu0
&
@rolf_comma_e
for their leadership.
#aiforgood
📣The arxiv proceedings of
#ML4D2020
are live!
🔗
Check out the great projects the ML4D community has been working on🎉
Please reach out to authors if you want to learn more about a specific project or to help them reach further success!
Excited about this collaboration and our proposal paper today at the
#ICML2021
@ClimateChangeAI
workshop. Check out our spotlight presentation at 9pm GMT+1! 🎊🤩
New papér at ICML CCAI 🚨
Satellites can overestimate how much carbon is stored by small reforestation projects by a factor of 10x. We need more trustworthy monitoring methods for forest carbon offsets at scale.
by
@GyriReiersen
,
@dwddao
, i,
@kklmmr
,
@xiaoxiang_zhu
, Ce Zhang
Check out how we produced high-resolution poverty maps in the Philippines by combining machine learning with open-sourced datasets.
We’ll be featured this Saturday at the 2020
@NeurIPSConf
@ML4Dworkshop
workshop!
Read our paper:
New paper out on improving graph neural networks (GNNs) for geospatial coordinates! Joint work with Daniel Neill and Nathan Safir.🤓🌍
#GeoAI
📰Paper:
💻Code:
🧵Thread: 👇
1/10
Happy to announce that I will be joining the
@turinginst
for the coming year of my PhD. I want to use my time there to enhance my knowledge of
#MachineLearning
and further examine the importance of spatial context in
#urban
problems
The
@ClimateChangeAI
Wiki team has been working really hard on this! Check out this new resource on all things climate change + AI & help us fill the gaps by contributing your expertise 💪
🥁Introducing: the
@ClimateChangeAI
Wiki, a community resource & knowledge collection for all topics at the intersection of
#ClimateChange
&
#MachineLearning
:
Eager to participate? Follow our guide & start contributing right away:
It’s a wrap🌯! Last Friday, we concluded this year’s edition of the
@ClimateChangeAI
summer school. A big thank you to everyone involved and our 2022 cohort. 🎉👏
Stay tuned for recordings of summer school content!
‼️Tomorrow is the big day!
@ML4Dworkshop
is returning to
#NeurIPS2021
with an amazing lineup of speakers and papers. Super excited for the
@ClimateChangeAI
workshop as well... it will be a full day 🤯🎊
#ML4D
2021 is excited to announce the interesting line up of our invited speakers
Date: Dec 14th; 2 p.m GMT.
Register for timely notifications about the event:
Join our mailing list:
New paper out in ACM
#IMWUT
today, joint with
@WarwickDCS
colleagues! We improve on existing methods for demand prediction and expansion planning in EV carsharing systems with spatio-temporal encoding and graph neural nets. (1/7)
➡️Paper:
⬇️Thread!
Catching up with a rapidly expanding field of deep learning for dynamical systems can be an intimidating task.
I am happy to share our preprint reviewing the latest methods in physics-informed neural networks (PINNs) and neural time steppers used for modeling dynamical systems.
Join the amazing team of
@ClimateChangeAI
🎉🎉. We're looking for volunteers in a range of different roles. It is a great, welcoming and ambitious atmosphere 💪
🌏CCAI is looking for new core team members in various volunteer roles, from event organization to media outreach! If you share our passion for
#ClimateChange
and
#ML
, head over to our website to learn more & apply:
One thing I forgot to add: Our Github features some interactive notebooks running on Google Colab, so you can try out our method straight away, without any need to download / clone the repo and without any personal GPU available!🔥🥳
Here is the link:
Very happy to present my new
#GeoML
🌏🛰️ paper: SXL, a framework for better learning with geographic data using spatially explicit auxiliary tasks
Joint work with Daniel Neill at
@NYU_CUSP
.
Paper:
Code:
Thread: 🔽
1/9
We've launched a new blog at
@ClimateChangeAI
for more casual reads & opinion sharing. Stay tuned for interesting interviews, perspectives, research summaries and more!
🥁We are excited to announce the
@ClimateChangeAI
blog!
On our blog, you'll find research summaries, perspectives, interviews and other exciting readings on various topics at the intersection of
#AI
&
#ClimateChange
. 🤓📚
Read on here:
#ICLR2020
is starting tomorrow! I will be giving two talks:
1⃣ April 26 17:30 GMT - ML-IRL workshop - Population estimation in informal settlements with equitable ground-truth
2⃣ April 27, 12:00 GMT - CCAI workshop - Demand prediction for EV sharing systems
We are super excited to announce that CCAI has partnered with
@IDRC_CRDI
,
@AI4Dev
&
@Sida
to support the funding of a new $1M innovation network on AI & climate action in Africa. The call for expressions of interest is now open:
Are you at
#NeurIPS2023
and interested in postdoc positions with
@MSFTResearch
? Join me on Dec 14 at 9.15am at the
@Microsoft
booth for a morning coffee ☕️and a postdoc Q&A session.
Full booth schedule is here: