This mathematical tool could improve how fast information is shared
advancedsciencenews.com/this-mathemati…
English
Torsten Ensslin
24 posts
Torsten Ensslin retweetledi
📣Just published: "Simulating Reputation Dynamics and Their Manipulation: An Agent Based Model Framework", with great thanks to the authors @TorstenEnsslin, Viktoria Kainz, and @CelineBoehm1. Available now: aup-online.com/content/journa…
English
Manipulative communication in humans and machines: Our work using agent-based simulations to understand certain socio-psychological phenomena, such as echo chambers, self-deception, camp formation, social moods, symbiosis of rogues, ...
mpa-garching.mpg.de/1075123/hl2023…
English
@RealJoshAlbert Yes, that is absolutely possible. One does not need deep networks to reconstruct Greens functions from data, see doi.org/10.1002/andp.2… .
English
Fantastic work on the way to having interpretable AI. @TorstenEnsslin I'm sure you know these people. WDYT about this work? marktechpost.com/2022/04/07/res…
English
@RealJoshAlbert No, that was new to me. Thanks for letting me know.
English
@amitseta90 We all needed to recharge afterwards. Congratulations again, Amit!
English
Three years to my PhD defence... (4.5 hours closed room viva but totally worth it)... @TorstenEnsslin (external) AnvarShukurov (primary supervisor) AndrewFletcher (mentor)...
And of course, we went for Indian food later to celebrate... :P
English
@bwandelt @thegautamkamath physicist's notation while stating the probability that is averaged over, in order not to contribute to the usual confusion in statistics: < X >_(X|A) := \sum_X P(X|A) X
English
Which of these do you prefer for writing the expectation of a random variable X?
In case Unicode doesn't work, 𝔼 is \mathbb{E}
English
@GameArtAcademic Sure. Sorry for the late response, I just saw your tweet now. What do you need beside nifty? Write me an email if you are still interested.
English
geoVI: Our novel information geometry based variational inference scheme for very high dimensional Bayesian data analysis problems, arxiv.org/abs/2105.10470 #DataScience
English
Sources of my last tweet:
US Covid-19 map from Nov 6th 2020:
#cases_casesper100klast7days" target="_blank" rel="nofollow noopener">covid.cdc.gov/covid-data-tra…
US election pattern from Nov 6th 2020:
#mapmode=lead" target="_blank" rel="nofollow noopener">edition.cnn.com/election/2020/…
English
Do US Corona virus spread and election pattern correlate?
English
𝙋𝙡𝙖𝙣𝙘𝙠 - 𝙈𝙞𝙨𝙨𝙞𝙤𝙣 𝘾𝙤𝙢𝙥𝙡𝙚𝙩𝙚𝙙
The Planck legacy publications appeared today:
aanda.org/articles/aa/ab…
Also, last Planck editorial board meeting today
For me: 16 years of collaboration work, 10 years editorial work, good science, and a very nice final picture:
English
Of harps, Christmas trees, a wandering star and the mysterious streams of cosmic rays
mpa-garching.mpg.de/786348/news201…
English
A smart comment on being over exited about having fitted some data in #AI or elsewhere:
Real numbers, data science and chaos:
How to fit any dataset with a single parameter
Laurent Boué
arxiv.org/abs/1904.12320
English
A small correction, the NIFTy5 documentation can (now) be found here:
ift.pages.mpcdf.de/nifty/
English