Log In Sign Up

Certification of Iterative Predictions in Bayesian Neural Networks

by   Matthew Wicker, et al.

We consider the problem of computing reach-avoid probabilities for iterative predictions made with Bayesian neural network (BNN) models. Specifically, we leverage bound propagation techniques and backward recursion to compute lower bounds for the probability that trajectories of the BNN model reach a given set of states while avoiding a set of unsafe states. We use the lower bounds in the context of control and reinforcement learning to provide safety certification for given control policies, as well as to synthesize control policies that improve the certification bounds. On a set of benchmarks, we demonstrate that our framework can be employed to certify policies over BNNs predictions for problems of more than 10 dimensions, and to effectively synthesize policies that significantly increase the lower bound on the satisfaction probability.


page 4

page 7

page 11

page 12

page 13

page 14


Modern Lower Bound Techniques in Database Theory and Constraint Satisfaction

Conditional lower bounds based on P≠ NP, the Exponential-Time Hypothesis...

Safety Guarantees for Planning Based on Iterative Gaussian Processes

Gaussian Processes (GPs) are widely employed in control and learning bec...

Lower Bounds on Sparse Spanners, Emulators, and Diameter-reducing shortcuts

We prove better lower bounds on additive spanners and emulators, which a...

On Lower Bounds for Regret in Reinforcement Learning

This is a brief technical note to clarify the state of lower bounds on r...

Robust Bayesian reinforcement learning through tight lower bounds

In the Bayesian approach to sequential decision making, exact calculatio...

Lower-bounds on the Bayesian Risk in Estimation Procedures via f-Divergences

We consider the problem of parameter estimation in a Bayesian setting an...