Jun, 2024
MaIL: 提升 Mamba 的模仿学习
MaIL: Improving Imitation Learning with Mamba
Xiaogang Jia, Qian Wang, Atalay Donat, Bowen Xing, Ge Li...
TL;DRMamba Imitation Learning (MaIL) is a versatile architecture that improves the performance of state space models and rivals against Transformers, outperforming them in all LIBERO tasks, achieving good performance with small datasets, effectively processing multi-modal sensory inputs, and being more robust to input noise.