Gunnar König

62 posts

Gunnar König banner
Gunnar König

Gunnar König

@gcskoenig

@gunnark.bsky.social (he/him) | PostDoc @uni_tue | prev. PhD @sldslmu & @univienna | #causality #xai

München, Bayern Katılım Mayıs 2013
523 Takip Edilen181 Takipçiler
Sabitlenmiş Tweet
Gunnar König
Gunnar König@gcskoenig·
I'm looking forward to presenting our work on "Improvement-Focused Causal Recourse (ICR)" at #AAAI23 today. Come by in 145A at 3:45 pm or come to poster #201 at 7 pm tonight! arxiv.org/abs/2210.15709
English
1
1
22
4K
Gunnar König retweetledi
Sebastian Bordt
Sebastian Bordt@sbordt·
New Blog Post!🎭
Sebastian Bordt tweet media
English
1
2
39
9.9K
Gunnar König retweetledi
Ulrike Luxburg
Ulrike Luxburg@UlrikeLuxburg·
I am looking for a PhD student in explainable machine learning. You need strong mathematical skills (the goal is to formulate theorems and prove them) and an open mind when it comes to talking to applied ML, social science, ethics or law. tinyurl.com/4ptp5amd
English
3
25
62
8K
Osman Ali Mian
Osman Ali Mian@osmanmian·
✅ Causal Discovery ✅ Federated Learning ✅ Privacy Guarantees Find out how we achieve this combo with nothing but regrets. 😄Today 1630hrs CEST in Auditorium 1 Foyer 71. #AISTATS2023 Joint work with David Kaltenpoth, @MichaelKamp7 and @drjilles.
Osman Ali Mian@osmanmian

At @aistats2023, we demonstrate how having regrets can be a good thing. 🙃 We show how you can learn #Causal Networks, in a #federated setting, with privacy #guarantees, without even needing to know the local models! 🤯 and to do so, regret is all you need! 😉

English
1
1
8
752
Gunnar König
Gunnar König@gcskoenig·
@thesephist Really cool, but I'd still prefer SO. The solution relies on np. If using np is allowed, one may also use np.random.normal to sample from the normal.
English
0
0
0
0
Linus
Linus@thesephist·
Made a little CLI that just pipes my programming questions to GPT-3, so I now can ask it stuff when I'm in the command line! LLMs are better than Stack Overflow now — I just ask it, and it gives me a comprehensive answer in one shot, right there in my terminal, in a couple secs.
Linus tweet media
English
72
337
3.4K
0
Peli Grietzer
Peli Grietzer@peligrietzer·
I feel like SJ and EA are the two cultural movements of my lifetime?
English
12
2
27
0
Gunnar König
Gunnar König@gcskoenig·
@brodriguesco @ChristophMolnar Models of observational distributions can help in causal inference, in general they do not solve the more complex task of causal inference. While for ML the goal is to predict (obs. dist), for causal inference the goal is to model the effect of interventions (int. distr).
English
0
0
1
0
Bruno Rodrigues
Bruno Rodrigues@brodriguesco·
@ChristophMolnar Now that I read your answer, I think we already had this discussion in the past :D what would you say about causal inference just being one task that can be solved using ml, freq or bayes?
English
1
0
0
0
Christoph Molnar 🦋 christophmolnar.bsky.social
Schools of data modeling - Machine learning - Frequentist statistics - Bayesian statistics - Causal inference Each school has its own basic assumptions about how to generalize from data to the world. Any other schools?
English
9
1
35
0
Gunnar König retweetledi
Joris Mooij
Joris Mooij@JorisMooij·
After having worked on this for about 8 years, I am very glad that our paper Foundations of Structural Causal Models with Cycles and Latent Variables has finally been published! Behind a paywall at projecteuclid.org/journals/annal… and openly accessible at arxiv.org/abs/1611.06221
English
3
43
247
0
Gunnar König retweetledi
Christoph Molnar 🦋 christophmolnar.bsky.social
Our paper on General Pitfalls of Model-Agnostic Interpretation Methods for Machine Learning Models is on arxiv! 🎉🎉🎉 arxiv.org/abs/2007.04131 These pitfalls can really mess up your model interpretation. Our paper can help you avoid them. A thread. 1/n 🧵
English
7
83
321
0
Gunnar König retweetledi
Moxie Marlinspike
Moxie Marlinspike@moxie·
Signal tried to use Instagram ads to display the data Facebook collects about you and sells access to. Facebook wasn't into the idea, and shut down our account instead: signal.org/blog/the-insta…
Moxie Marlinspike tweet media
English
152
5.6K
19K
0
Gunnar König retweetledi
uai2026
uai2026@UncertaintyInAI·
Because of problems with prolonged power outages in multiple countries, #UAI2021 deadline is extended to 20 February 23:59 UTC. We hope that this short extension is of some help to those who have been experiencing difficulties. Please retweet! auai.org/uai2021/
English
1
18
29
0
Gunnar König
Gunnar König@gcskoenig·
Are you a student who cares about the long-term impact of AI on our societies and would like to help shape the field? Then consider applying for the #DataScience for #SocialGood summer school. Closing date: Jan 31 @datascifellows @turinginst @sldslmu
SLDS / Bernd Bischl@BBischl

Together with @warwickuni we’re organising the @DSSGxUK Summer Projects 2021 fellowship. If you’re an undergrad or postgrad student with a passion for #DataScience and #SocialGood, apply NOW. bit.ly/DSSGx2021 Closing date: Jan. 31 @datascifellows @turinginst @sldslmu

English
0
2
4
0
tino ⏸
tino ⏸@bomelino·
sag mir dein problem und ich mal was dazu was nicht hilft
Deutsch
130
46
1K
0