Simulation Priors for Data-Efficient Deep Learning
This addresses the data-efficiency problem for AI systems in complex real-world environments like biology, agriculture, and robotics, representing a novel method for a known bottleneck.
The authors tackled the problem of enabling AI systems to learn efficiently in real-world settings by proposing SimPEL, a method that combines first-principles models with data-driven learning using low-fidelity simulators as priors in Bayesian deep learning, resulting in superior performance in learning complex dynamics and learning a high-speed RC car parking maneuver with substantially less data than state-of-the-art baselines.
How do we enable AI systems to efficiently learn in the real-world? First-principles models are widely used to simulate natural systems, but often fail to capture real-world complexity due to simplifying assumptions. In contrast, deep learning approaches can estimate complex dynamics with minimal assumptions but require large, representative datasets. We propose SimPEL, a method that efficiently combines first-principles models with data-driven learning by using low-fidelity simulators as priors in Bayesian deep learning. This enables SimPEL to benefit from simulator knowledge in low-data regimes and leverage deep learning's flexibility when more data is available, all the while carefully quantifying epistemic uncertainty. We evaluate SimPEL on diverse systems, including biological, agricultural, and robotic domains, showing superior performance in learning complex dynamics. For decision-making, we demonstrate that SimPEL bridges the sim-to-real gap in model-based reinforcement learning. On a high-speed RC car task, SimPEL learns a highly dynamic parking maneuver involving drifting with substantially less data than state-of-the-art baselines. These results highlight the potential of SimPEL for data-efficient learning and control in complex real-world environments.