Michał Bortkiewicz
@michalbortkiewicz.bsky.social
PhD @ Warsaw University of Technology
JacGCRL
Reinforcement Learning / Continual Learning / Neural Networks Plasticity
JacGCRL
Reinforcement Learning / Continual Learning / Neural Networks Plasticity
🚨Scaling RL
Most RL methods’ performance saturate at ~5 layers. In this work led by Kevin Wang, we crack the right configuration for scaling Contrastive RL and go beyond 1000 layers NNs! Deep NNs unlock emergent behaviors and other cool properties. Check out Kevin’s thread!
Most RL methods’ performance saturate at ~5 layers. In this work led by Kevin Wang, we crack the right configuration for scaling Contrastive RL and go beyond 1000 layers NNs! Deep NNs unlock emergent behaviors and other cool properties. Check out Kevin’s thread!
1/ While most RL methods use shallow MLPs (~2–5 layers), we show that scaling up to 1000-layers for contrastive RL (CRL) can significantly boost performance, ranging from doubling performance to 50x on a diverse suite of robotic tasks.
Webpage+Paper+Code: wang-kevin3290.github.io/scaling-crl/
Webpage+Paper+Code: wang-kevin3290.github.io/scaling-crl/
March 20, 2025 at 11:22 PM
🚨Scaling RL
Most RL methods’ performance saturate at ~5 layers. In this work led by Kevin Wang, we crack the right configuration for scaling Contrastive RL and go beyond 1000 layers NNs! Deep NNs unlock emergent behaviors and other cool properties. Check out Kevin’s thread!
Most RL methods’ performance saturate at ~5 layers. In this work led by Kevin Wang, we crack the right configuration for scaling Contrastive RL and go beyond 1000 layers NNs! Deep NNs unlock emergent behaviors and other cool properties. Check out Kevin’s thread!
Reposted by Michał Bortkiewicz
🚀 What happens when you modify the spectrum of singular values of the merged task vector? 🤔
Apparently, you achieve 🚨state-of-the-art🚨 model merging results! 🔥
✨ Introducing “No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces”
Apparently, you achieve 🚨state-of-the-art🚨 model merging results! 🔥
✨ Introducing “No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces”
February 10, 2025 at 2:47 PM
🚀 What happens when you modify the spectrum of singular values of the merged task vector? 🤔
Apparently, you achieve 🚨state-of-the-art🚨 model merging results! 🔥
✨ Introducing “No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces”
Apparently, you achieve 🚨state-of-the-art🚨 model merging results! 🔥
✨ Introducing “No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces”
Reposted by Michał Bortkiewicz
Reinforcement learning agents should be able to improve upon behaviors seen during training.
In practice, RL agents often struggle to generalize to new long-horizon behaviors.
Our new paper studies *horizon generalization*, the degree to which RL algorithms generalize to reaching distant goals. 1/
In practice, RL agents often struggle to generalize to new long-horizon behaviors.
Our new paper studies *horizon generalization*, the degree to which RL algorithms generalize to reaching distant goals. 1/
February 4, 2025 at 8:37 PM
Reinforcement learning agents should be able to improve upon behaviors seen during training.
In practice, RL agents often struggle to generalize to new long-horizon behaviors.
Our new paper studies *horizon generalization*, the degree to which RL algorithms generalize to reaching distant goals. 1/
In practice, RL agents often struggle to generalize to new long-horizon behaviors.
Our new paper studies *horizon generalization*, the degree to which RL algorithms generalize to reaching distant goals. 1/
Just hit 100 stars on GitHub with JaxGCRL! ✨
We're working on a frictionless experience so users can discover, install, and run their first experiment in under 10 minutes.
Want to join the team? We're looking for contributors to make JaxGCRL the go-to GCRL repository! 🚀
github.com/MichalBortki...
We're working on a frictionless experience so users can discover, install, and run their first experiment in under 10 minutes.
Want to join the team? We're looking for contributors to make JaxGCRL the go-to GCRL repository! 🚀
github.com/MichalBortki...
December 16, 2024 at 7:23 PM
Just hit 100 stars on GitHub with JaxGCRL! ✨
We're working on a frictionless experience so users can discover, install, and run their first experiment in under 10 minutes.
Want to join the team? We're looking for contributors to make JaxGCRL the go-to GCRL repository! 🚀
github.com/MichalBortki...
We're working on a frictionless experience so users can discover, install, and run their first experiment in under 10 minutes.
Want to join the team? We're looking for contributors to make JaxGCRL the go-to GCRL repository! 🚀
github.com/MichalBortki...
Reposted by Michał Bortkiewicz
For my first post on Bluesky .. I'll start by announcing our 2025 edition of EEML which will be in Sarajevo :) ! I'm really excited about it and hope to see many of you there. Please follow the website (and Bluesky account) for more details which are coming soon ..
Hello Bluesky! 🦋
This will be the official account of the Eastern European Machine Learning (EEML) community.
Follow us for news regarding our summer schools, workshops, education/community initiatives, and more!
This will be the official account of the Eastern European Machine Learning (EEML) community.
Follow us for news regarding our summer schools, workshops, education/community initiatives, and more!
December 15, 2024 at 6:39 PM
For my first post on Bluesky .. I'll start by announcing our 2025 edition of EEML which will be in Sarajevo :) ! I'm really excited about it and hope to see many of you there. Please follow the website (and Bluesky account) for more details which are coming soon ..
Reposted by Michał Bortkiewicz
HOT 🔥 fastest, most precise, and most capable hand control setup ever...
Less than $450 and fully open-source 🤯
by @huggingface, @therobotstudio, @NepYope
This tendon-driven technology will disrupt robotics! Retweet to accelerate its democratization 🚀
A thread 🧵
Less than $450 and fully open-source 🤯
by @huggingface, @therobotstudio, @NepYope
This tendon-driven technology will disrupt robotics! Retweet to accelerate its democratization 🚀
A thread 🧵
December 15, 2024 at 8:22 AM
HOT 🔥 fastest, most precise, and most capable hand control setup ever...
Less than $450 and fully open-source 🤯
by @huggingface, @therobotstudio, @NepYope
This tendon-driven technology will disrupt robotics! Retweet to accelerate its democratization 🚀
A thread 🧵
Less than $450 and fully open-source 🤯
by @huggingface, @therobotstudio, @NepYope
This tendon-driven technology will disrupt robotics! Retweet to accelerate its democratization 🚀
A thread 🧵
Reposted by Michał Bortkiewicz
🤖 Introducing RL Zero 🤖: a new approach to transform language into behavior zero-shot for embodied agents without labeled datasets!
December 11, 2024 at 7:11 AM
🤖 Introducing RL Zero 🤖: a new approach to transform language into behavior zero-shot for embodied agents without labeled datasets!
Reposted by Michał Bortkiewicz
Excited to invite you to our #NeurIPS spotlight poster "BRO: Bigger, Regularized, Optimistic"! 🎉
📍 Poster #6302
📅 West Ballroom A-D
🕚 Friday, 11:00-14:00
Join us to discuss with Michał Nauman and me. Let’s talk SOTA in RL! 💪
🧵👇
📍 Poster #6302
📅 West Ballroom A-D
🕚 Friday, 11:00-14:00
Join us to discuss with Michał Nauman and me. Let’s talk SOTA in RL! 💪
🧵👇
December 12, 2024 at 3:58 PM
Excited to invite you to our #NeurIPS spotlight poster "BRO: Bigger, Regularized, Optimistic"! 🎉
📍 Poster #6302
📅 West Ballroom A-D
🕚 Friday, 11:00-14:00
Join us to discuss with Michał Nauman and me. Let’s talk SOTA in RL! 💪
🧵👇
📍 Poster #6302
📅 West Ballroom A-D
🕚 Friday, 11:00-14:00
Join us to discuss with Michał Nauman and me. Let’s talk SOTA in RL! 💪
🧵👇