Model-based Validation as Probabilistic Inference
This work addresses validation challenges for autonomous systems, offering a more efficient and comprehensive method for failure estimation, though it is incremental as it builds on existing Bayesian and model-based techniques.
The authors tackled the problem of estimating failure distributions for autonomous systems by framing it as Bayesian inference, using model-based rollouts and automatic differentiation to compute trajectory gradients, and demonstrated improved sample efficiency and parameter space coverage in experiments with an inverted pendulum, autonomous vehicle, and lunar lander.
Estimating the distribution over failures is a key step in validating autonomous systems. Existing approaches focus on finding failures for a small range of initial conditions or make restrictive assumptions about the properties of the system under test. We frame estimating the distribution over failure trajectories for sequential systems as Bayesian inference. Our model-based approach represents the distribution over failure trajectories using rollouts of system dynamics and computes trajectory gradients using automatic differentiation. Our approach is demonstrated in an inverted pendulum control system, an autonomous vehicle driving scenario, and a partially observable lunar lander. Sampling is performed using an off-the-shelf implementation of Hamiltonian Monte Carlo with multiple chains to capture multimodality and gradient smoothing for safe trajectories. In all experiments, we observed improvements in sample efficiency and parameter space coverage compared to black-box baseline approaches. This work is open sourced.