Matthias Pirlet

PhD student @UniversiteLiege. Interested by: AI, electricity markets, MLOps

[Gaspard Lambrechts @GsprdLambrechts]

Yet another asymmetric RL algo: Asymmetric Advantage Weighted Regression (AAWR).

This time, the goal is to learn a policy pi(a | z) whose input z = f(h) is not necessarily Markovian. It is useful in robotics, for example.

So, how to adapt RL in POMDP to non Markovian input?

🧵

[Damien ERNST @DamienERNST1]

Annual dinner of my research group at the University of Liège .
I am so full of admiration for all these young clever people who want to push further the boundaries of knowledge for a better world.
Here is a link to all our scientific publications:
https://t.co/isccBAfmek

Many thanks also to Marc Carnevale from Les Sabots d'Hélène for welcoming us so well in its restaurant.

On the picture : Edouard Princen, Pascal Leroy, Maurizio Vassallo, Lize Pirenne, Raphael Fonteneau, Bardhyl Miftari, Céline Metz, Antoine Mouchamps, Antoine Larbanois, Victor Dachet, Anthony Maio, Julien Hansen, Thibaut Techy, Ahmed Tayachi, Guillaume Derval, Thomas Richard, Julien Di Renzo, Arthur Malherbe, Adrien Bolland, Matthias Pirlet, Pierre Counotte, Loris Bigatton, Julien Brandoit, Gaspard Lambrechts, Antoine Mouchamps,
Samy Aittahar, Martin Schoors, Guy Langenaeken, Titah Khaled Raouf
Oussama Chaouch, Robin Gaspari,
Elias Ernst, Alireza Bahmanyar, Arthur Louette, Louis Colson, Damien Ernst, Samy Mokeddem plus of course Marc Carnevale.

[François Rozet @FrancoisRozet]

📢 I am looking for AI post-doc/research/engineer positions in Europe (Paris, London, Zurich, ...). My work revolves around generative modeling and AI for Science, with 4+ publications at top conferences during my PhD. If you are hiring, please reach out! If not, please repost 🔁

[sacha lewin @SachaLewin]

We introduce Appa 🦬, a 1.5B-parameter probabilistic large weather model capable of various downstreak tasks without retraining!

This collaborative effort with my whole team was a real honor, and I'm glad to finally show it today!

[Adrien Bolland @AdrienBolland]

Check our work on max entropy RL! We introduce an off-policy method to maximize the entropy of future state-action visitation distributions, leading to policies that explore effectively and achieve high performance 🎯

Link 📑https://t.co/KEIv7kvSlp

#RL #MaxEntRL #Exploration

[vmoens @VincentMoens]

Internship offer!

We’re looking for an intern (research scientist/PhD) to join the #PyTorch team at Meta this summer.

The work involves GPU kernel generation using LLM 👇

[François Rozet @FrancoisRozet]

I'll be in Vancouver from Dec 9 to Dec 16 to present our work at NeurIPS 2024 🇨🇦 If you want to talk about generative models, ML for science, open-source software, or bouldering, send me a DM or an email!

[François Rozet @FrancoisRozet]

chat, what do we think about my poster?

[Damien ERNST @DamienERNST1]

Annual dinner of my research group.
I am so full of admiration for all these young clever people who want to push further the boundaries of knowledge for a better world.
Here is a link to all our scientific publications:
https://t.co/wXPURCIh2m

[Damien ERNST @DamienERNST1]

▪︎ A pleasure to welcome in Liège the great RL researcher Florian Felten 😀 We brought him for dinner at a typical Belgian restaurant.

▪︎ He will talk tomorrow at 11am at the Montefiore Institute about multi-objective reinforcement learning.
More information about his talk:
https://t.co/110OvbwMcQ

▪︎ From left to right on the picture: @DachetV , Samy Mokeddem,
@MatthiasPirlet, Maurizio Vassallo, Pascal Leroy , @FlorianFelten1 and @DamienERNST1 .

[Florian Felten @FlorianFelten1]

For what it's worth; the slides of my talk at ULiège are available here.

[Matthias Pirlet @MatthiasPirlet]

@AdrienBolland @glouppe @DamienERNST1 7/7 📖 You can read the full paper here: https://t.co/rEgmhlnFPr

[Matthias Pirlet @MatthiasPirlet]

1/7 📝 Excited to share that my paper, "Cost Estimation in Unit Commitment Problems Using Simulation-Based Inference," has been accepted at the D3S3 workshop at NeurIPS! Here's what it's about 👇

[Matthias Pirlet @MatthiasPirlet]

6/7 I’m deeply grateful to the GMA team at Engie, especially Alexandre Huynen for his insightful discussions. Thanks also to @AdrienBolland for his detailed reviews, @glouppe for his SBI expertise, and @DamienERNST1 for his guidance and support throughout this project.

[Sasha Rush @srush_nlp]

Updating all my NeurIPS papers.

[Matthias Pirlet @MatthiasPirlet]

7/7 📖 You can read the full paper here: https://t.co/rEgmhlnFPr