PhD Student in the group of @JanHasenauer @UniBonn
Bonn, Deutschland
Joined September 2018
- Tweets58
- Following149
- Followers50
- Likes270
Probably one of the most stunning results of our = @JanHasenauer|s research group:
Federated learning has a HUGHE security issue: Data can recovered completely, only using the covariance function, which basically all federated learning frameworks have!
biorxiv.org/content/10.1101/…
Jakob Vanhoefer retweeted
🎒Summer school: 𝐈𝐧𝐯𝐞𝐫𝐬𝐞 𝐩𝐫𝐨𝐛𝐥𝐞𝐦𝐬 𝐟𝐨𝐫 𝐦𝐮𝐥𝐭𝐢-𝐬𝐜𝐚𝐥𝐞 𝐦𝐨𝐝𝐞𝐥𝐬
💬With: Linda Petzold, Christiane Fuchs, @dennisprangle, @StefanEngblom, @A_Hellander
⏰August 22-26
📌@HCM_Bonn
✏️Details+register: hcm.uni-bonn.de/events/event…
Wir = LightningPose freuen uns wirklich riesig! Vielen Dank und wir sind gespannt, wie sich das Projekt weiter entwickelt!
Been to one of @cthoyt's amazing introductions to python package design and it really shaped the way I write (scientific) software.
I can really recommend them (July 25, 5-8PM CEST), if you want to learn about turning python code to a clean pip-installable python package!
Our collaborator and friend @cthoyt is generously hosting a free webinar on designing reproducible projects with Python. This may be useful to many. Do join if your schedule permits. More info at eventbrite.de/e/reusable-rep…
Really happy to see my first first-author paper on #yaml2sbml out.
Ever had an ODE, that you wanted to simulate/fit in your standard #SBML-pipeline and wondered, how get the model to #SBML? 🚀
Just published in #JOSS_theOJ: 'yaml2sbml: Human-readable and -writable specification of ODE models and their conversion to SBML' doi.org/10.21105/joss.03215
Last but not least: Thank you a lot to the @JOSS_TheOJ, especially @arfon, @kelly_l_rowland and our reviewers @SirSharpest and @marouenbg for their effort and helpful suggestions that improved the paper and lead to a really pleasant reviewing experience! (Crazy, right?)🚀
Jakob Vanhoefer retweeted
Well-written piece of code for converting ODE models, comes with docs, CI, tutorials. Totally recommend 👍
Just published in #JOSS_theOJ: 'yaml2sbml: Human-readable and -writable specification of ODE models and their conversion to SBML' doi.org/10.21105/joss.03215
Looking forward to present our work on easy specification of #ODE-models in #SBML with #yaml2sbml at #HARMONY2021 later today.
github.com/yaml2sbml-dev/yam…
Only 5 days left until #HARMONY2021 starts. Join us for lightning talks, hacking, community sessions and a bit of vision about the fufure of #COMBINE. We're looking forward to over 130 participants! co.mbine.org/events/HARMONY_…
Jakob Vanhoefer retweeted
Interested in #CompBio, #SingleCells or #MathModelling with applications in #metainflammation or #epidemiology?
Check out our open PhD and PostDoc positions:
nature.com/naturecareers/job…
nature.com/naturecareers/job…
Partners: @open_econ, @orchestracohort
github.com/ICB-DCM/pyPESTO
#pyPESTO v0.2.3 is out & comes with lots of new improvements, e.g.:
- Two new optimizers interfaced (NLopt and @fabfrohlich's wonderful FIDES)
- import of #PySB models
- extended #PEtab support
- predictions from fitted models
- unconstrained optimization
- Extended storage and visualization routines.
- and many more...
Big shoutout to @yannik_schaelte @fabfrohlich @dweindl @StaporPaul @PLakrisenko, Dilan, Elba, Paul Jost and Phillipp (all not on twitter) for contributing to this release!
Looking forward to present #pyPESTO, a #python package for parameter estimation & uncertainty quantification at this year's #COMBINE2020.
Thank you to the organizers for giving me the opportunity to present!
The schedule for this year's #COMBINE2020 is ready. Check out who's speaking, who's offering tutorials, and what topics are going to be discussed in breakouts - and when: co.mbine.org/sites/default/f… Registration is still possible through the website co.mbine.org/events/COMBINE_…
Happy to present #pyPESTO a #python package for parameter estimation & uncertainty quantification at this year's #COMBINE2020.
You can find the jupyter notebook containing all the code (plus some bonus material) and the slides here:
github.com/jvanhoefer/COMBIN…
Jakob Vanhoefer retweeted
#pyPESTO 0.1.0 is out! 🎉😃
Parameter estimation toolbox featuring (esp.):
- efficient gradient-based optimization
- sampling with @pymc_devs and parallel tempering
- profiling
- result visualization
- integrated with #AMICI and #PEtab
➡️ github.com/icb-dcm/pypesto
