Bertrand Charpentier
@bertrand-sharp.bsky.social
Founder, President & Chief Scientist @PrunaAI | Prev. @Twitter research, Ph.D. in ML @TU_Muenchen
Alibaba models
- replicate.com/wan-video/wa...
- replicate.com/wan-video/wa...
- replicate.com/prunaai/vace...
- replicate.com/prunaai/wan-...
- replicate.com/qwen/qwen-im...
- replicate.com/wan-video/wa...
- replicate.com/wan-video/wa...
- replicate.com/prunaai/vace...
- replicate.com/prunaai/wan-...
- replicate.com/qwen/qwen-im...
August 6, 2025 at 1:29 PM
HiDream models:
- replicate.com/prunaai/hidr...
- replicate.com/prunaai/hidr...
- replicate.com/prunaai/hidr...
- replicate.com/prunaai/hidr...
- replicate.com/prunaai/hidr...
- replicate.com/prunaai/hidr...
- replicate.com/prunaai/hidr...
- replicate.com/prunaai/hidr...
August 6, 2025 at 1:29 PM
Flux models:
- replicate.com/prunaai/flux...
- replicate.com/prunaai/flux...
- replicate.com/prunaai/flux...
- replicate.com/prunaai/flux...
- replicate.com/prunaai/flux...
- replicate.com/prunaai/flux...
August 6, 2025 at 1:29 PM
Hope you will enjoy the paper and the code 🙂
Paper: arxiv.org/pdf/2402.15978
Code: github.com/fortuinlab/s...
Paper: arxiv.org/pdf/2402.15978
Code: github.com/fortuinlab/s...
arxiv.org
December 6, 2024 at 4:29 PM
Hope you will enjoy the paper and the code 🙂
Paper: arxiv.org/pdf/2402.15978
Code: github.com/fortuinlab/s...
Paper: arxiv.org/pdf/2402.15978
Code: github.com/fortuinlab/s...
In this paper 📄
- We train sparsifiable models via the Marginal Likelihood to make pruning easy after training ✅
- We construct parameter- & unit-wise prior that fit unstructured and structured pruning ✅
- We propose the cheap Optimal Posterior Damage score outperforming other pruning scores ✅
- We train sparsifiable models via the Marginal Likelihood to make pruning easy after training ✅
- We construct parameter- & unit-wise prior that fit unstructured and structured pruning ✅
- We propose the cheap Optimal Posterior Damage score outperforming other pruning scores ✅
December 6, 2024 at 4:29 PM
In this paper 📄
- We train sparsifiable models via the Marginal Likelihood to make pruning easy after training ✅
- We construct parameter- & unit-wise prior that fit unstructured and structured pruning ✅
- We propose the cheap Optimal Posterior Damage score outperforming other pruning scores ✅
- We train sparsifiable models via the Marginal Likelihood to make pruning easy after training ✅
- We construct parameter- & unit-wise prior that fit unstructured and structured pruning ✅
- We propose the cheap Optimal Posterior Damage score outperforming other pruning scores ✅
Reposted by Bertrand Charpentier
This was a super fun collaboration, led by my master's student Rayen Dhahri with Alex Immer,
@bertrand-sharp.bsky.social, and Stephan Günnemann
@bertrand-sharp.bsky.social, and Stephan Günnemann
December 6, 2024 at 3:05 PM
This was a super fun collaboration, led by my master's student Rayen Dhahri with Alex Immer,
@bertrand-sharp.bsky.social, and Stephan Günnemann
@bertrand-sharp.bsky.social, and Stephan Günnemann
Hope you will enjoy the paper and the overview 🙂
Paper: openreview.net/pdf/b84db14f...
Overview: www.cs.cit.tum.de/daml/expecte...
4/4
Paper: openreview.net/pdf/b84db14f...
Overview: www.cs.cit.tum.de/daml/expecte...
4/4
openreview.net
December 6, 2024 at 1:58 PM
Hope you will enjoy the paper and the overview 🙂
Paper: openreview.net/pdf/b84db14f...
Overview: www.cs.cit.tum.de/daml/expecte...
4/4
Paper: openreview.net/pdf/b84db14f...
Overview: www.cs.cit.tum.de/daml/expecte...
4/4
It is part of a long research on structure learning :)
- "Differentiable DAG Sampling", ICML 2022
- "End-to-end Learning of Probabilistic Hierarchies on Graphs", ICLR 2021
- "Tree Sampling Divergence", IJCAI 2019
- "Hierarchical graph clustering using node pair sampling" MLG, KDD 2018
3/4
- "Differentiable DAG Sampling", ICML 2022
- "End-to-end Learning of Probabilistic Hierarchies on Graphs", ICLR 2021
- "Tree Sampling Divergence", IJCAI 2019
- "Hierarchical graph clustering using node pair sampling" MLG, KDD 2018
3/4
End-to-end learning of probabilistic hierarchies on graphs
D Zügner, B Charpentier, M Ayle, S Geringer, S Günnemann, International Conference on Learning Representations, 2021 - 3-mal zitiert
scholar.google.com
December 6, 2024 at 1:58 PM
It is part of a long research on structure learning :)
- "Differentiable DAG Sampling", ICML 2022
- "End-to-end Learning of Probabilistic Hierarchies on Graphs", ICLR 2021
- "Tree Sampling Divergence", IJCAI 2019
- "Hierarchical graph clustering using node pair sampling" MLG, KDD 2018
3/4
- "Differentiable DAG Sampling", ICML 2022
- "End-to-end Learning of Probabilistic Hierarchies on Graphs", ICLR 2021
- "Tree Sampling Divergence", IJCAI 2019
- "Hierarchical graph clustering using node pair sampling" MLG, KDD 2018
3/4
In this paper 📄
- We optimize the expected objectives over probabilistic model on hierarchies. ✅
- We differentiably sample any hierarchy (not only binary ones!). ✅
- We show theoretical consistency and achieve SOTA results for hierarchical learning. ✅
2/4
- We optimize the expected objectives over probabilistic model on hierarchies. ✅
- We differentiably sample any hierarchy (not only binary ones!). ✅
- We show theoretical consistency and achieve SOTA results for hierarchical learning. ✅
2/4
December 6, 2024 at 1:58 PM
In this paper 📄
- We optimize the expected objectives over probabilistic model on hierarchies. ✅
- We differentiably sample any hierarchy (not only binary ones!). ✅
- We show theoretical consistency and achieve SOTA results for hierarchical learning. ✅
2/4
- We optimize the expected objectives over probabilistic model on hierarchies. ✅
- We differentiably sample any hierarchy (not only binary ones!). ✅
- We show theoretical consistency and achieve SOTA results for hierarchical learning. ✅
2/4
🚀 𝗟𝗼𝗼𝗸𝗶𝗻𝗴 𝗔𝗵𝗲𝗮𝗱
Our journey continues as we grow our team of 𝘁𝗼𝗽 𝘁𝗮𝗹𝗲𝗻𝘁𝘀 𝗮𝗰𝗿𝗼𝘀𝘀 𝗘𝘂𝗿𝗼𝗽𝗲 🇪🇺, with hubs in Germany and France 🇩🇪🤝🇫🇷. Together, we’re enabling cutting-edge AI compression and deployment without compromising real-world performance.
Our journey continues as we grow our team of 𝘁𝗼𝗽 𝘁𝗮𝗹𝗲𝗻𝘁𝘀 𝗮𝗰𝗿𝗼𝘀𝘀 𝗘𝘂𝗿𝗼𝗽𝗲 🇪🇺, with hubs in Germany and France 🇩🇪🤝🇫🇷. Together, we’re enabling cutting-edge AI compression and deployment without compromising real-world performance.
November 27, 2024 at 6:28 PM
🚀 𝗟𝗼𝗼𝗸𝗶𝗻𝗴 𝗔𝗵𝗲𝗮𝗱
Our journey continues as we grow our team of 𝘁𝗼𝗽 𝘁𝗮𝗹𝗲𝗻𝘁𝘀 𝗮𝗰𝗿𝗼𝘀𝘀 𝗘𝘂𝗿𝗼𝗽𝗲 🇪🇺, with hubs in Germany and France 🇩🇪🤝🇫🇷. Together, we’re enabling cutting-edge AI compression and deployment without compromising real-world performance.
Our journey continues as we grow our team of 𝘁𝗼𝗽 𝘁𝗮𝗹𝗲𝗻𝘁𝘀 𝗮𝗰𝗿𝗼𝘀𝘀 𝗘𝘂𝗿𝗼𝗽𝗲 🇪🇺, with hubs in Germany and France 🇩🇪🤝🇫🇷. Together, we’re enabling cutting-edge AI compression and deployment without compromising real-world performance.
🌟 𝗟𝗼𝗼𝗸𝗶𝗻𝗴 𝗕𝗮𝗰𝗸
What started as an idea in my PhD, became an incredible journey. Bringing together my brilliant co-founders -S. Günnemann, J. Rachwan, R. Nait Mazi- was a pivotal step for Pruna AI. In 1 year, we smashed thousands of AI models, making them 𝗰𝗵𝗲𝗮𝗽𝗲𝗿💲, 𝗳𝗮𝘀𝘁𝗲𝗿⚡, 𝘀𝗺𝗮𝗹𝗹𝗲𝗿💾, and 𝗴𝗿𝗲𝗲𝗻𝗲𝗿🌳.
What started as an idea in my PhD, became an incredible journey. Bringing together my brilliant co-founders -S. Günnemann, J. Rachwan, R. Nait Mazi- was a pivotal step for Pruna AI. In 1 year, we smashed thousands of AI models, making them 𝗰𝗵𝗲𝗮𝗽𝗲𝗿💲, 𝗳𝗮𝘀𝘁𝗲𝗿⚡, 𝘀𝗺𝗮𝗹𝗹𝗲𝗿💾, and 𝗴𝗿𝗲𝗲𝗻𝗲𝗿🌳.
November 27, 2024 at 6:28 PM
🌟 𝗟𝗼𝗼𝗸𝗶𝗻𝗴 𝗕𝗮𝗰𝗸
What started as an idea in my PhD, became an incredible journey. Bringing together my brilliant co-founders -S. Günnemann, J. Rachwan, R. Nait Mazi- was a pivotal step for Pruna AI. In 1 year, we smashed thousands of AI models, making them 𝗰𝗵𝗲𝗮𝗽𝗲𝗿💲, 𝗳𝗮𝘀𝘁𝗲𝗿⚡, 𝘀𝗺𝗮𝗹𝗹𝗲𝗿💾, and 𝗴𝗿𝗲𝗲𝗻𝗲𝗿🌳.
What started as an idea in my PhD, became an incredible journey. Bringing together my brilliant co-founders -S. Günnemann, J. Rachwan, R. Nait Mazi- was a pivotal step for Pruna AI. In 1 year, we smashed thousands of AI models, making them 𝗰𝗵𝗲𝗮𝗽𝗲𝗿💲, 𝗳𝗮𝘀𝘁𝗲𝗿⚡, 𝘀𝗺𝗮𝗹𝗹𝗲𝗿💾, and 𝗴𝗿𝗲𝗲𝗻𝗲𝗿🌳.
I would be happy to be part of it :)
November 21, 2024 at 8:39 PM
I would be happy to be part of it :)
That looks great! I would love to be in :)
November 20, 2024 at 11:02 AM
That looks great! I would love to be in :)