Michał Bortkiewicz
@m_bortkiewicz
Followers
149
Following
569
Media
9
Statuses
93
PhD at Warsaw University of Technology Working on RL and Continual Learning
Joined February 2021
I am excited to share our recent work with @WladekPalucki , @vivek_myers, @Taddziarm , @tomArczewski, @LukeKucinski, and @ben_eysenbach!. Accelerating Goal-Conditioned Reinforcement Learning Algorithms and Research . Webpage:
1
23
83
RT @sukjun_hwang: Tokenization has been the final barrier to truly end-to-end language models. We developed the H-Net: a hierarchical netw….
0
718
0
RT @e7mul: 1/7 If Andrew Ng is right that the LR is the most important ML hyperparam, it's got some competition! We show that the softmax t….
0
12
0
RT @jan_dubinski_: 🚨We’re thrilled to present our paper “CDI: Copyrighted Data Identification in #DiffusionModels” at #CVPR2025 in Nashvill….
0
7
0
RT @chongyiz1: 1/ How should RL agents prepare to solve new tasks? While prior methods often learn a model that predicts the immediate next….
0
33
0
RT @jonathanrichens: Are world models necessary to achieve human-level agents, or is there a model-free short-cut?.Our new #ICML2025 paper….
0
175
0
RT @GhugareRaj: Normalizing Flows (NFs) check all the boxes for RL: exact likelihoods (imitation learning), efficient sampling (real-time c….
0
18
0
RT @kvfrans: Stare at policy improvement and diffusion guidance, and you may notice a suspicious similarity. We lay out an equivalence b….
0
46
0
RT @mic_nau: We wondered if off-policy RL could transfer to real robots on-par with on-policy PPO. Turns out it works surprisingly well!….
0
9
0
RT @bartoszcyw: New paper: Deceptive LLMs may keep secrets from their operators. Can we elicit this latent knowledge? Maybe!. Our LLM knows….
0
19
0
RT @AndrewZ45732491: ❄️Introducing Absolute Zero Reasoner: Our reasoner learns to both propose tasks that maximize learnability and improve….
0
341
0
RT @piotrsankowski: Mixture of Experts (MoE) one of the solutions used by DeepSeek. In MoE LLMs only part of the model parameters is activa….
0
10
0
RT @axlewandowski: I will be presenting two posters at ICLR that outlines an optimization perspective on loss of plasticity. Come check the….
0
5
0
Excited to present JaxGCRL at ICLR 2025 (spotlight):. 📍Hall 3 + Hall 2B, Poster #422.🗓️Friday, April 25.🕒3:00 PM – 5:00 PM . I'm also happy to grab a coffee and chat about anything related to RL, robotics, or continual learning!.
I am excited to share our recent work with @WladekPalucki , @vivek_myers, @Taddziarm , @tomArczewski, @LukeKucinski, and @ben_eysenbach!. Accelerating Goal-Conditioned Reinforcement Learning Algorithms and Research . Webpage:
0
3
20
RT @tomasztrzcinsk1: Lukasz @LukeKucinski is not only an excellent researcher, but a truely great person. Kind, thoughtful and wise. And al….
0
3
0
RT @PiotrRMilos: My good friend has an ongoing fight with cancer. A great father and husband for his family. An excellent co-author for me….
0
13
0
RT @piotrsankowski: Instytut Ideas może zacząć działać - wczoraj dokonany został wpis do KRSu. Jednakże wśród dobrych informacji są też tak….
0
84
0
RT @ben_eysenbach: tldr: increase the depth of your RL networks by several orders of magnitude. Our new paper shows that very very deep ne….
0
31
0
RT @kevin_wang3290: 1/ While most RL methods use shallow MLPs (~2–5 layers), we show that scaling up to 1000-layers for contrastive RL (CRL….
0
66
0
RT @IJ_Apps: Check out this new paper by @kevin_wang3290, myself, @m_bortkiewicz, @tomasztrzcinsk1, and.@ben_eysenbach! .We show a method f….
0
1
0
🚨Scaling RL.Most RL methods’ performance saturate at ~5 layers. In this work led by @kevin_wang3290, 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:
0
7
24