
Researcher @UNPeacekeeping | Research Associate RiskEcon Lab NYU | Ex @UNGlobalPulse | PhD in Physics @IPPP_Durham. Married to @CAylettBullock .
New York City
Joined October 2012
- Tweets198
- Following1.1K
- Followers293
- Likes1.1K
Great to see that the ‘Principles for the Ethical Use of #AI in the @UN System’ have now been published - it was a pleasure to work on this with many colleagues across the UN chaired by @ITU and @UNESCO unsceb.org/principles-ethica…
Joseph Aylett-Bullock retweeted
Great development in the German funding landscape !
🚨Game changer alert🚨 for scientists in Germany who publish and maintain #opensource software and #opendata. Our @github and @ZENODO_ORG etc repositories will be considered as publications when applying for funding @dfg_public 🤩🚀🤯
Joseph Aylett-Bullock retweeted
Multi-AI Complex Systems in Humanitarian Response
deepai.org/publication/multi…
by @JosephPBullock et al.
#ArtificialIntelligence #Computers
Joseph Aylett-Bullock retweeted
Today UN Global Pulse's @josephpbullock and @litonidas will share about our work on Multi-AI Complex Systems in Humanitarian Response. #KDD2022 @HumanitarianSys
Learn more here: kdd-humanitarian-mapping.her…
Joseph Aylett-Bullock retweeted
Today is World Creativity & Innovation Day!
How do you define #innovation?
Not everyone knows that you can help protect #refugees & displaced populations by applying epidemic modelling, data & team work to fight the pandemic.
This is innovation. bit.ly/3rgXLMc
#WCID
Joseph Aylett-Bullock retweeted
From pandemics to conflict, human rights must be respected and humanitarian assistance should use the best tools at our disposal. Proud of colleagues working on and writing this practice piece to improve epidemic response for people fleeing their homes. gh.bmj.com/content/bmjgh/7/3…
Joseph Aylett-Bullock retweeted
Our AI & Data Science Research Fellow @YennieJun takes a deep dive into our work with @WHO on scaling 📻 radio data analysis with #DataScience for #infodemic monitoring:
unglobalpulse.org/2022/01/sc…
#ASR #NLP #COVID19
Happy to see the official press release of our research since the beginning of the pandemic working with some amazing people @NHSuk @ucl to support the @royalsociety modelling initiate to better understand the spread of #COVID19
Experts from @DurhamPhysics have developed a new model to simulate the spread of Covid-19 in collaboration with @NHSEngland and @ucl. Sounds interesting?
Find out more 👉 fal.cn/3itz4
Great to be presenting today @ DLR #HumTechDays event on our work modelling the spread of COVID-19 in refugee and IDP settlements in collaboration w/ @durham_uni @UNGlobalPulse @humdata @MITIBMLab @LSHTM @arnauquera @aidans203 @hen_str @litonidas @rebecamoreno @UNHCRInnovation
Experts have gathered online for @DLR_de’s #HumTechDays 2021! 🌏 On this year's agenda - how #technology can help meet humanitarian needs resulting from the COVID-19 pandemic 😷 & how to reach the @UN #Sustainable Development Goals by 2030. More ℹ️: dlr.expert/htd2021/front/ind…
More information on our work can be found here: unglobalpulse.org/microsite/…
Joseph Aylett-Bullock retweeted
Experts have gathered online for @DLR_de’s #HumTechDays 2021! 🌏 On this year's agenda - how #technology can help meet humanitarian needs resulting from the COVID-19 pandemic 😷 & how to reach the @UN #Sustainable Development Goals by 2030. More ℹ️: dlr.expert/htd2021/front/ind…
Our new #ML in #HEP paper is out - and it’s the first one under my new name! We use NNs to approximate functions which are computationally intensive to evaluate and speed up particle physics simulations for the discovery of new phenomena
Optimising simulations for diphoton production at hadron colliders using amplitude neural networks
deepai.org/publication/optim…
by @JosephPBullock et al.
#NeuralNetwork #MachineLearning
Joseph Aylett-Bullock retweeted
In the final session on the Ethics & Society stage at #CogX2021 today, we're delighted to be joined by @mmitchell_ai, co-founder of Google's Ethical AI team, to talk about how we ensure technology is developed in a safe, responsible & meaningful way in the next 10 years.
Really timely and important work by @SashaMTL and @josephdviviano - especially as we utilise language models more and more we can’t just use large datasets for training simply because they’re large without a clear understanding of their content
Probably not at all surprising, but there's a lot of sketchy content online and it's being used to train language models. @josephdviviano and I did our best to quantify exactly how much there is!
Joseph Aylett-Bullock retweeted
Probably not at all surprising, but there's a lot of sketchy content online and it's being used to train language models. @josephdviviano and I did our best to quantify exactly how much there is!
Always brilliant @SashaMTL and I looked into the darkest recesses of the @CommonCrawl to find what unsavory content large language models are trained on, and didn't like what we found! Our work was accepted into @aclmeeting and you can read about it here: arxiv.org/abs/2105.02732
Joseph Aylett-Bullock retweeted
For just over a year of the COVID-19 pandemic, we highlighted key papers and preprints as they appeared, to help readers keep up with the flood of coronavirus research. Here's that archive: go.nature.com/3oUmm5v
Thanks for hosting a great first session @WHO @TDPurnat! - pleasure to present our work and looking forward to the next session
Tune in as our own @JosephPBullock presents the public #radio data analysis tool we're developing to support @WHO response to the #COVID19 infodemic:
Via the WHO 4th Infodemic Management Conference
youtube.com/watch?v=VTRJr5HQ…
Joseph Aylett-Bullock retweeted
Tune in as our own @JosephPBullock presents the public #radio data analysis tool we're developing to support @WHO response to the #COVID19 infodemic:
Via the WHO 4th Infodemic Management Conference
youtube.com/watch?v=VTRJr5HQ…
Joseph Aylett-Bullock retweeted
Diving into the epidemic models that can teach us about the spread of disease (#COVID19 ) and intervention planning. Here's how we used computer simulations in refugee settlements. A collab w/ @UNHCRInnovation, @WHO, @MITIBMLab & others.
epimodel.unglobalpulse.net/
























