AdaptDiffuser: Diffusion Models as Adaptive Self-evolving Planners
25. července 2023
Řečníci
Zhixuan Liang
Řečník · 0 sledujících
Mingyu Ding
Řečník · 1 sledující
Masayoshi Tomizuka
Řečník · 0 sledujících
O prezentaci
Diffusion models have demonstrated their powerful generative capability in many tasks, with great potential to serve as a paradigm for offline reinforcement learning. However, the quality of the diffusion model is limited by the insufficient diversity of training data, which hinders the performance of planning and the generalizability to new tasks. This paper introduces AdaptDiffuser, an evolutionary planning method with diffusion that can self-evolve to improve the diffusion model hence a better planner, not only for seen tasks but can also adapt to unseen tasks. AdaptDiffuser enables the generation of rich synthetic expert data for goal-conditioned tasks using guidance from reward gradients. It then selects high-quality data via a discriminator to finetune the diffusion model, which improves the generalization ability to unseen tasks. Empirical experiments on two benchmark environments and two carefully designed unseen tasks in KUKA industrial robot arm and Maze2D environments demonstrate the effectiveness of AdaptDiffuser. For example, AdaptDiffuser not only outperforms the previous art Diffuser by 20.8% on Maze2D and 7.5% on MuJoCo locomotion, but also adapts better to new tasks, e.g., KUKA pick-and-place, by 27.9% without requiring additional expert data. More visualization results and demo videos could be found on our project page.Diffusion models have demonstrated their powerful generative capability in many tasks, with great potential to serve as a paradigm for offline reinforcement learning. However, the quality of the diffusion model is limited by the insufficient diversity of training data, which hinders the performance of planning and the generalizability to new tasks. This paper introduces AdaptDiffuser, an evolutionary planning method with diffusion that can self-evolve to improve the diffusion model hence a bette…
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
Anytime Model Selection in Linear Bandits
Zhlédnout později
Zeroth-Order Optimization Meets Human Feedback: Provable Learning via Ranking Oracles
Zhlédnout později
Efficient Learning of Mesh-Based Physical Simulation with Bi-Stride Multi-Scale Graph Neural Network
Zhlédnout později
Yadi Cao, …
SWARM Parallelism: Training Large Models Can Be Surprisingly Communication-Efficient
Zhlédnout později
Max Ryabinin, …
Online Platt Scaling with Calibeating
Zhlédnout později
Chirag Gupta, …
Practical and Matching Gradient Variance Bounds for Black-Box Variational Bayesian Inference
Zhlédnout později
Kyurae Kim, …