EfficientFlow: Efficient Equivariant Flow Policy Learning for Embodied AI
This addresses data and sampling inefficiency in embodied AI policy learning, representing an incremental improvement over existing generative methods.
The paper tackles the data inefficiency and slow inference of generative policies in embodied AI by introducing EfficientFlow, which incorporates equivariance into flow matching and uses acceleration regularization. The method achieves competitive or superior performance on robotic manipulation benchmarks with limited data while providing dramatically faster inference.
Generative modeling has recently shown remarkable promise for visuomotor policy learning, enabling flexible and expressive control across diverse embodied AI tasks. However, existing generative policies often struggle with data inefficiency, requiring large-scale demonstrations, and sampling inefficiency, incurring slow action generation during inference. We introduce EfficientFlow, a unified framework for efficient embodied AI with flow-based policy learning. To enhance data efficiency, we bring equivariance into flow matching. We theoretically prove that when using an isotropic Gaussian prior and an equivariant velocity prediction network, the resulting action distribution remains equivariant, leading to improved generalization and substantially reduced data demands. To accelerate sampling, we propose a novel acceleration regularization strategy. As direct computation of acceleration is intractable for marginal flow trajectories, we derive a novel surrogate loss that enables stable and scalable training using only conditional trajectories. Across a wide range of robotic manipulation benchmarks, the proposed algorithm achieves competitive or superior performance under limited data while offering dramatically faster inference. These results highlight EfficientFlow as a powerful and efficient paradigm for high-performance embodied AI.