NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration (Summary Video) @rail7462
NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration (Summary Video)  @rail7462
Uploaded October 2023 | Updated November 2026, 7 hours ago
We present Navigation with Goal Masking and Diffusion (NoMaD), a Transformer-based policy trained on data from multiple ground robots, with a diffusion model decoder to flexibly handle both goal-conditioned and goal-agnostic navigation.

"NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration"
Ajay Sridhar, Dhruv Shah, Catherine Glossop, Sergey Levine
Berkeley AI Research (UC Berkeley)

Project Page: general-navigation-models.github.io
arXiv pre-print: arxiv.org/abs/2310.07896
NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration (Summary Video)CS 285: Lecture 22, Part 3: Transfer Learning & Meta-LearningCS 185/285 (Spring 2026): Lecture 24, Multi-Task and Hierarchical RLCS 182: Lecture 2, Part 3: Machine Learning BasicsReinforcement Learning Pretraining for Reinforcement Learning FinetuningData-Driven Reinforcement Learning for Robotic Manipulation
RAIL |

NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration (Summary Video)

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER