Walter Mayor-Toro

65 posts

Walter Mayor-Toro banner
Walter Mayor-Toro

Walter Mayor-Toro

@WalterMayor_T

Katılım Mayıs 2019
48 Takip Edilen44 Takipçiler
Sabitlenmiş Tweet
Walter Mayor-Toro
Walter Mayor-Toro@WalterMayor_T·
1/13
Today, I’m thrilled to share a major milestone—
A dream sparked in a small town of farms and chalkboards 🌾📚
My first paper as an independent researcher was accepted to one of the world’s top AI & ML conferences.
Here’s how I got here. 👇 arxiv.org/pdf/2506.03404
English
3
0
38
3.8K
Walter Mayor-Toro retweetledi
Pablo Samuel Castro
Pablo Samuel Castro@pcastr·
🚨The Formalism-Implementation Gap in RL research🚨 Lots of progress in RL research over last 10 years, but too much performance-driven => overfitting to benchmarks (like the ALE). 1⃣ Let's advance science of RL 2⃣ Let's be explicit about how benchmarks map to formalism 1/X
Pablo Samuel Castro tweet media
English
2
28
154
9.4K
Walter Mayor-Toro
Walter Mayor-Toro@WalterMayor_T·
Proud to have contributed, as an independent researcher, to the great work achieved by our research team. Many thanks to @johanobandoc , @lavoiems , Scott Fujimoto, @AaronCourville and @pcastr for making it happen. See Pablo’s thread for more details.
Pablo Samuel Castro@pcastr

🔊Simplicial Embeddings (SEMs) Improve Sample Efficiency in Actor-Critic Agents🔊 In our recent preprint we demonstrate that the use of well-structured representations (SEMs) can dramatically improve sample efficiency in RL agents. 1/X

English
1
4
9
1.4K
Walter Mayor-Toro
Walter Mayor-Toro@WalterMayor_T·
Huge thanks to all the amazing researchers who made this possible, and especially to @johanobandoc for his guidance and for welcoming me, as an independent researcher, into this and many other projects.
It’s been a truly enriching and fascinating journey
Johan Obando-Ceron 👍🏽@johanobandoc

1/3 🥳Excited to share our new paper ‘Simplicial Embeddings Improve Sample Efficiency in Actor–Critic Agents’! Project your features onto a product of simplices → sparse, stable reps, stronger grads, faster learning. 🧵For more details, check out Pablo’s thread 👇

English
0
1
4
474
Walter Mayor-Toro retweetledi
Johan Obando-Ceron 👍🏽
Johan Obando-Ceron 👍🏽@johanobandoc·
🚨Super excited to share that our paper “Stable Gradients for Stable Learning at Scale in DRL” was accepted at #NeurIPS as a Spotlight! ✨🌟 Paper: arxiv.org/abs/2506.15544 Huge thanks to my amazing coauthors. I’ve learned so much from each of you. 🫶 Onward! 🚀 #NeurIPS2025
Johan Obando-Ceron 👍🏽 tweet media
Roger Creus Castanyer@creus_roger

Scaling up networks in deep RL is tricky: do it naively, and performance collapses 😵‍💫 Why? Increasing depth and width leads to severe vanishing gradients, causing unstable learning ⚠️ We diagnose this issue across several algorithms: DQN, Rainbow, PQN, PPO, SAC, and DDPG.

English
3
23
91
13.2K
Walter Mayor-Toro
Walter Mayor-Toro@WalterMayor_T·
13/13
 If you also come from a place where people say “it can’t be done”
 Here’s my message: 💥 Yes, it can.
 🌾 There’s science in our villages.
 🏙️ There’s talent in our neighborhoods.
 🚀 And there’s a future for those who dream big.
English
1
0
9
247
Walter Mayor-Toro
Walter Mayor-Toro@WalterMayor_T·
12/13 I hope this paper marks the beginning of an amazing and passionate journey as a researcher, one where I can learn, create, and collaborate with many talented researchers.
GIF
English
1
0
5
294
Walter Mayor-Toro
Walter Mayor-Toro@WalterMayor_T·
1/13
Today, I’m thrilled to share a major milestone—
A dream sparked in a small town of farms and chalkboards 🌾📚
My first paper as an independent researcher was accepted to one of the world’s top AI & ML conferences.
Here’s how I got here. 👇 arxiv.org/pdf/2506.03404
English
3
0
38
3.8K

Keşfet

@lavoiems @johanobandoc @AaronCourville @pcastr @__Vaibhavi @Mila_Quebec @Theo_Vincent_ @_LXAI