GoTube: Scalable Stochastic Verification of Continuous-Depth Models

Abstract

We introduce a new stochastic verification algorithm that formally quantifies the behavioural robustness of any time-continuous process formulated as a continuousdepth model. The algorithm solves a set of global optimization (Go) problems over a given time horizon to construct a tight enclosure (Tube) of the set of all process executions starting from a ball of initial states. We call our algorithm GoTube. Through its construction, GoTube ensures that the bounding tube is conservative up to a desired probability. GoTube is implemented in JAX and optimized to scale to complex continuous-depth models. Compared to advanced reachability analysis tools for time-continuous neural networks, GoTube provably does not accumulate over-approximation errors between time steps, and avoids the infamous wrapping effect inherent in symbolic techniques. We show that GoTube substantially outperforms state-of-the-art verification tools in terms of the size of the initial ball, speed, time-horizon, task completion, and scalability, on a large set of experiments. GoTube is stable and sets the state-of-the-art for its ability to scale up to time-horizons well-beyond what has been possible before.

Differentiable Control Barrier Functions for Vision-based End-to-End Autonomous Driving

“…OptNet with CBFs has been used in neural networks as a filter for safe controls [36], but OptNet is not trainable, thus, potentially limiting the system's learning performance. In [14,25,59,16,17], safety guaranteed neural network controllers have been learned through verification-in-the-loop training. A safe neural network filter has been proposed in [15] for a specific vehicle model using verification methods.…”

BarrierNet: A Safety-Guaranteed Layer for Neural Networks

“…Optimization-based safety frameworks. Recent advances in differentiable optimization methods show promising directions for safety guaranteed neural network controllers (Lechner et al, 2020b;Gruenbacher et al, 2020;Grunbacher et al, 2021;Gruenbacher et al, 2021;Massiani et al, 2021;. In (Amos and Kolter, 2017), a differentiable quadratic program (QP) layer, called OptNet, was introduced.…”