Antoine Leeman

880 posts

Antoine Leeman

Antoine Leeman

@antoine_leeman

Robotics PhD @eth_en @esa | prev. @Rai_int @MIT | optimization, ML | opinions = mine

Zurich, Switzerland Katılım Nisan 2012
1.3K Takip Edilen1.3K Takipçiler
Sabitlenmiş Tweet
Antoine Leeman
Antoine Leeman@antoine_leeman·
Yann Lecun favours MPC over RL? Interesting! What about data-augmented MPC? @ylecun #nmpc2024
Antoine Leeman tweet media
English
20
41
372
340.8K
Antoine Leeman retweetledi
Tom Silver
Tom Silver@tomssilver·
This week's #PaperILike is "Differentiable GPU-Parallelized Task and Motion Planning" (Shen et al., RSS 2025). As always, meticulous work from @WillShenSaysHi and team. TAMP + GPU is long overdue! PDF: arxiv.org/abs/2411.11833
English
0
12
103
13K
Yung-Sung Chuang
Yung-Sung Chuang@YungSungChuang·
🎓 Life updates: I defended my PhD last week at MIT. Now I am joining OpenAI to continue my research on building trustworthy LLMs! Thanks my advisor James Glass and my committee @yoonrkim, @jacobandreas and all my friends for continual supports throughout this journey!
Yung-Sung Chuang tweet mediaYung-Sung Chuang tweet mediaYung-Sung Chuang tweet media
English
50
47
1.1K
62.2K
Antoine Leeman retweetledi
NeurIPS Europe Conference
NeurIPS Europe Conference@EurIPSConf·
We are off to a great start with our first set of 9 #EurIPS workshops! the lecture rooms are buzzing and conversations are flowing at our three different workshop locations!
NeurIPS Europe Conference tweet mediaNeurIPS Europe Conference tweet mediaNeurIPS Europe Conference tweet mediaNeurIPS Europe Conference tweet media
English
0
4
13
1.1K
Antoine Leeman
Antoine Leeman@antoine_leeman·
Uncertainties in Machine Learning! Looking forward to this workshop! @EurIPSConf
Antoine Leeman tweet media
English
2
3
12
758
Antoine Leeman retweetledi
Yann LeCun
Yann LeCun@ylecun·
@rao2z Ok: - chain rule existed. - Lagrangian mechanics existed. - the adjoint state methods in optimal control existed. But no one had shown chain rule could be used to train multilayer nets. And *clearly* *no one* had actually made it work until about 1985.
English
18
13
279
0
Antoine Leeman retweetledi
NeurIPS Europe Conference
NeurIPS Europe Conference@EurIPSConf·
Our last keynote at #EurIPS is happening in just a few minutes! We will welcome @GaelVaroquaux to the podium for his talk "Uncertainty in the LLM era - Science, more than scale"
NeurIPS Europe Conference tweet media
English
0
1
5
408
Antoine Leeman retweetledi
NeurIPS Europe Conference
NeurIPS Europe Conference@EurIPSConf·
We are 7 hours into today's program, but we are not done yet! In just a few minutes Michael Jordan will be giving today's second #EurIPS keynote on "A Collectivist, Economic Perspective on AI". So load up on coffee and see you there! ☕
NeurIPS Europe Conference tweet media
English
0
1
9
562
Antoine Leeman retweetledi
NeurIPS Europe Conference
NeurIPS Europe Conference@EurIPSConf·
We hope you all enjoyed the lunch break! Our first #EurIPS keynote today by @HochreiterSepp is happening in just a few minutes! We are looking forward to his talk on "Sustainable, Low-Energy, and Fast AI Made in Europe"!
NeurIPS Europe Conference tweet media
English
0
3
13
721
Antoine Leeman retweetledi
NeurIPS Europe Conference
NeurIPS Europe Conference@EurIPSConf·
Hope you enjoy our first #EurIPS poster session! Remember to go talk to our sponsors and all of your peers!
NeurIPS Europe Conference tweet mediaNeurIPS Europe Conference tweet mediaNeurIPS Europe Conference tweet media
English
0
1
9
550
Antoine Leeman retweetledi
Antoine Leeman retweetledi
NeurIPS Conference
NeurIPS Conference@NeurIPSConf·
NeurIPS is pleased to officially endorse EurIPS, an independently-organized meeting taking place in Copenhagen this year, which will offer researchers an opportunity to additionally present their accepted NeurIPS work in Europe, concurrently with NeurIPS. Read more in our blog post and on the EurIPS website: blog.neurips.cc/2025/07/16/neu… eurips.cc
English
10
120
796
116.8K
Antoine Leeman retweetledi
Physical Intelligence
Physical Intelligence@physical_int·
We train a general-purpose value function on all of our data, which tells the π*0.6 VLA which actions are good or bad. By asking π*0.6 to produce only good actions, we get better performance. We call this method Recap.
Physical Intelligence tweet media
English
3
3
82
16K
Antoine Leeman
Antoine Leeman@antoine_leeman·
Excited to be at #EurIPS this week, the European venue for #NeurIPS-accepted papers 🎓 Working on predictive control + machine learning for robotics 🤖 If you’re into #RobotLearning or control, let’s connect!
Antoine Leeman tweet media
English
0
0
2
206