Pinned
Excited to share Flow Matching Policy Gradients: expressive RL policies trained from rewards using flow matching. It’s an easy, drop-in replacement for Gaussian PPO on control tasks.
00:00
David McAllister
192 posts
- Decentralized Diffusion Models power stronger models trained on more accessible infrastructure. DDMs mitigate the networking bottleneck that locks training into expensive and power-hungry centralized clusters. They scale gracefully to billions of parameters and generate
00:00 - Replying to @davidrmcallCheck out our blog post at flowreinforce.github.io We developed interactive plots that explain the connection between flow/diffusion models and RL. w/ a great team of collaborators! @Songwei_Ge @brenthyi @ChungMinKim @ethanjohnweber @redstone_hong @HavenFeng @akanazawa
- Replying to @davidrmcall @Songwei_Ge and 6 othersPaper link:
- I’ll be at #NeurIPS2024 this week presenting Rethinking Score Distillation as a Bridge Between Image Distributions! Poster Presentation: Friday 4:30-7:30 PM Come chat with me or @holynski_ about lifting diffusion models to 3D!
00:00 - Humanoids on campus! Check out our new work on context-aware locomotionour new system trains humanoid robots using data from cell phone videos, enabling skills such as climbing stairs and sitting on chairs in a single policy (w/ @redstone_hong @junyi42 @davidrmcall)
00:00 - @akanazawa and I will be presenting Decentralized Diffusion Models tomorrow (Friday) morning at CVPR poster session 1, come find us!Decentralized Diffusion Models power stronger models trained on more accessible infrastructure. DDMs mitigate the networking bottleneck that locks training into expensive and power-hungry centralized clusters. They scale gracefully to billions of parameters and generate00:00
- RDM is now published at Nature Methods! This was a 3 year effort and my introduction to academic research. I’m fortunate to have been mentored by one of the smartest people I’ve ever met @the_legitamit!Ring deconvolution microscopy is now published at @naturemethods! nature.com/articles/s4159… There are some fun new additions including light-sheet deconvolution 🫡 Stay tuned for the official python package release next week! Any feature suggestions are more than welcome 😃
- Replying to @davidrmcallPlease see our paper for more details or blog for intuitive and interactive explanations. Blog: decentralizeddiffusion.github.io Paper:
- Replying to @davidrmcallStandard diffusion models communicate gradients at every optimization step, a network load that only centralized clusters can support. Decentralized diffusion models divide training into independent pieces that can proceed on different hardware in different locations. This is a
00:00 - Anything is possible with Viser!July has been a big month for Viser! - Released v1.0.0😊 - We did some writing Some demos👇
00:00 - Replying to @davidrmcallThis was only possible thanks to mentors Matthew Tancik and Jiaming Song (@baaadas), as well as the support of the rest of the fantastic @LumaLabsAI research team! Thank you for hosting me as an intern.
- Replying to @davidrmcallThis produces an ensemble whose predictions combine at test-time. You can drop experts at test-time to save on computation. In fact, we only inference a single expert per step in our comparisons and we outperform standard diffusion models for with the same FLOP budgets at train
00:00
