Infinite Time Horizon Safety of Bayesian Neural Networks
Bayesian neural networks (BNNs) place distributions over the weights of a neural network to model uncertainty in the data and the network's prediction. We consider the problem of verifying safety when running a Bayesian neural network policy in a feedback loop with infinite time horizon systems. Compared to the existing sampling-based approaches, which are inapplicable to the infinite time horizon setting, we train a separate deterministic neural network that serves as an infinite time horizon safety certificate. In particular, we show that the certificate network guarantees the safety of the system over a subset of the BNN weight posterior's support. Our method first computes a safe weight set and then alters the BNN's weight posterior to reject samples outside this set. Moreover, we show how to extend our approach to a safe-exploration reinforcement learning setting, in order to avoid unsafe trajectories during the training of the policy. We evaluate our approach on a series of reinforcement learning benchmarks, including non-Lyapunovian safety specifications.
Code (1)
Tasks
reinforcement-learningReinforcement LearningReinforcement Learning (RL)Safe ExplorationSimilar Papers 제목 키워드 기반
Inverse Optimal Safety Filters
CBF-QP safety filters are pointwise minimizers of the control effort at a given state vector, i.e., myopically optimal at each time instant. But are they optimal over the entire infinite time horizon? What does it even m…
Of Good Demons and Bad Angels: Guaranteeing Safe Control under Finite Precision
As neural networks (NNs) become increasingly prevalent in safety-critical neural network-controlled cyber-physical systems (NNCSs), formally guaranteeing their safety becomes crucial. For these systems, safety must be en…
CVaR-based Safety Analysis in the Infinite Time Horizon Setting
We develop a risk-averse safety analysis method for stochastic systems on discrete infinite time horizons. Our method quantifies the notion of risk for a control system in terms of the severity of a harmful random outcom…
ManagementEfficient Online Learning with Offline Datasets for Infinite Horizon MDPs: A Bayesian Approach
In this paper, we study the problem of efficient online reinforcement learning in the infinite horizon setting when there is an offline dataset to start with. We assume that the offline dataset is generated by an expert …
Imitation LearningProvably Efficient Exploration in Constrained Reinforcement Learning:Posterior Sampling Is All You Need
We present a new algorithm based on posterior sampling for learning in constrained Markov decision processes (CMDP) in the infinite-horizon undiscounted setting. The algorithm achieves near-optimal regret bounds while be…
AllEfficient Exploration