LudwigBald Profile Banner
Ludwig Bald Profile
Ludwig Bald

@LudwigBald

Followers
285
Following
8K
Statuses
4K

Fixing the world with data, machine learning, cognitive science, politics, humane technology and effective altruism Machine Learning Engineer @bmz_bund

🇪🇺/Berlin
Joined February 2014
Don't wanna be here? Send us removal request.
@LudwigBald
Ludwig Bald
3 years
This is a diary thread where I will reply with meta master's thesis events and feels. Let's go!
2
0
6
@LudwigBald
Ludwig Bald
3 days
@MarkovMagnifico @matsonj I though this was a great measured overview of the state of AI coding, seen on hacker news:
0
0
2
@LudwigBald
Ludwig Bald
8 days
@catehall Party games should be super easy to explain and fun for all players. A few recent favorites are: - Insider - Top Ten - Just One
0
0
2
@LudwigBald
Ludwig Bald
9 days
@davidpomerenke @williameijer I recently saw stats saying that female refugees in Germany even have way fewer friends than male refugees. That seems bad!
0
0
1
@LudwigBald
Ludwig Bald
13 days
@emollick Or maybe they really want to accelerate tech and then at some point switch to closed models like you said. They're a big enough player that they would capture a lot of the public good that open models provide. Enough to justify the investment.
0
0
0
@LudwigBald
Ludwig Bald
14 days
@SquireDigital @kvistp Finnish is really different. There's a podcast where a Swede, a Dane and a Norwegian all speak their native language:
0
0
1
@LudwigBald
Ludwig Bald
15 days
@MariusHobbhahn Congratulations, Dr. Hobbhahn!
0
0
0
@LudwigBald
Ludwig Bald
17 days
RT @_alice_evans: Europe's falling fertility means our workforce can only grow through immigration. But are we supporting women's integra…
0
20
0
@LudwigBald
Ludwig Bald
18 days
@kareem_carr There's no such thing as an exact level of uncertainty, right? Because you don't know how certain you can be about your fundamental assumptions.
3
0
9
@LudwigBald
Ludwig Bald
1 month
@auderdy you can add "before:2022" to your search, that filters out the new stuff
1
0
3
@LudwigBald
Ludwig Bald
1 month
@paularambles gross!
0
0
1
@LudwigBald
Ludwig Bald
1 month
@S1R1US_BLVCK @Rainmaker1973 It's the air pressure pushing from outside. Vacuums can't pull, they just push less than normal air. Even a vacuum cleaner doesn't technically suck, it just makes some space available inside and then the outside air pushes in.
4
0
0
@LudwigBald
Ludwig Bald
1 month
@AviSchiffmann It doesn't make sense to focus on pets. Pets, and zoo animals, have a much better life than the billions of factory farmed animals! There are so many farmed animals. If you care about animal welfare, that's where your energy should go.
0
0
3
@LudwigBald
Ludwig Bald
1 month
@simon_ohler I think I saw someone post about the ministries' data labs, the idea stuck around, I applied the normal way.
0
0
2
@LudwigBald
Ludwig Bald
1 month
@andy_matuschak -1, because when that's a prime suddenly all whole numbers except 0 have prime factors. But in the spirit of the question, 51 has no business being a multiple of something
0
0
2
@LudwigBald
Ludwig Bald
1 month
@Simon__Grimm I mean it's not super liquid, you're welcome to bet! I agree with you. I think the CDU-Greens coalition will be possible mathematically, but deep down parts of the CDU really don't want to work with the Greens, so maybe they will just use them to gain leverage over the SPD.
0
0
1
@LudwigBald
Ludwig Bald
1 month
@Simon__Grimm Nice, here it is! They simulate outcomes based on polls, they show pollsters' historical bias. But they don't try to predict effects of singular events, like Nate Silver does for e.g. party conventions. It seems like the polls will at most be a few % off
0
0
1
@LudwigBald
Ludwig Bald
1 month
@NathanpmYoung It's a fundamental problem with capitalism. Those with capital (in this example, China. Sometimes it's billionaires etc.) wield significant power over those with less capital. They get to influence the entire economy to serve them instead of consumers.
0
0
0