arXiv research

Unified control theory and machine learning for safety in uncertain systems.

Reinforcement Learning (RL) algorithms have found limited success beyond simulated applications, and one main reason is the absence of safety guarantees during the learning process. Real world systems would realistically fail or break before an optimal controller can be learned. To address this issue, we propose a cont…

Safe learning of stochastic dynamics with safety constraints.

Safe RL in linear systems achieves $\sqrt{T}$-regret.

Safety filter for unknown discrete-time systems with learned models and noise covariance.

Safe Bayesian optimization tackles safety constraints in control engineering.

SECRM-2D improves RL-based autonomous driving with safety guarantees.

Paper proposes Vertex Networks for reinforcement learning of control systems with safety guarantees.

This work establishes safe reinforcement learning for LQR with nonlinear baselines.

In order to satisfy safety conditions, an agent may be constrained from acting freely. A safe controller can be designed a priori if an environment is well understood, but not when learning is employed. In particular, reinforcement learned (RL) controllers require exploration, which can be hazardous in safety critical …

Deriving the optimal safety stock quantity with which to meet customer satisfaction is one of the most important topics in stock management. However, it is difficult to control the stock management of correlated marketable merchandise when using an inventory control method that was developed under the assumption that t…

We study the constrained linear quadratic regulator with unknown dynamics, addressing the tension between safety and exploration in data-driven control techniques. We present a framework which allows for system identification through persistent excitation, while maintaining safety by guaranteeing the satisfaction of st…

Develops probabilistic safety regions for scalable classifiers.

Recent successes in reinforcement learning have lead to the development of complex controllers for real-world robots. As these robots are deployed in safety-critical applications and interact with humans, it becomes critical to ensure safety in order to avoid causing harm. A first step in this direction is to test the …

Simple online monitor detects unsafe LLM outputs.

Proposes a recursive MPC scheme with probabilistic safety guarantees for uncertain dynamic systems.

This work improves motion planning for quadcopters by learning and reasoning about controller performance.

A novel controller for wheeled robots handles joystick inputs for smooth steering.

Meta-learning priors improves safe Bayesian optimization.

Failures are challenging for learning to control physical systems since they risk damage, time-consuming resets, and often provide little gradient information. Adding safety constraints to exploration typically requires a lot of prior knowledge and domain expertise. We present a safety measure which implicitly captures…

The paper shows why AI safety doesn't generalize across tasks.

A framework integrates machine learning with robust control for safer, more reliable systems.

Safe reinforcement learning framework using optimal transport for robustness.

Classifier-based AI safety gates fail in self-improvement, even with advanced verification methods.

CitySim dataset captures vehicle trajectories for safety research.

Machine and reinforcement learning (RL) are increasingly being applied to plan and control the behavior of autonomous systems interacting with the physical world. Examples include self-driving vehicles, distributed sensor networks, and agile robots. However, when machine learning is to be applied in these new settings,…

Iterative method learns unknown constraints for MPC control.

Combines Lyapunov functions with controller synthesis for safe control policies.

Deep learning has become an increasingly common technique for various control problems, such as robotic arm manipulation, robot navigation, and autonomous vehicles. However, the downside of using deep neural networks to learn control policies is their opaque nature and the difficulties of validating their safety. As th…

Safe RL-based vibration control using LQR guidance.

We study the problem of safe learning and exploration in sequential control problems. The goal is to safely collect data samples from operating in an environment, in order to learn to achieve a challenging control goal (e.g., an agile maneuver close to a boundary). A central challenge in this setting is how to quantify…

This paper explores formal verification for autonomous systems, identifying limitations and proposing improvements.

Safe learning in uncertain systems with state measurements and optimization.

A new protocol corrects confounding effects to measure alignment-induced activation shifts accurately.

Variable speed limits (VSL) control is a flexible way to improve traffic condition,increase safety and reduce emission. There is an emerging trend of using reinforcement learning technique for VSL control and recent studies have shown promising results. Currently, deep learning is enabling reinforcement learning to dev…

Data-driven models are subject to model errors due to limited and noisy training data. Key to the application of such models in safety-critical domains is the quantification of their model error. Gaussian processes provide such a measure and uniform error bounds have been derived, which allow safe control based on thes…

Safe linear bandits over unknown polytopes avoid safety violations and suboptimal actions.

Safe exploration method for RL under disturbance ensures safety with probabilistic guarantees.

ARTEO algorithm optimizes safety-critical systems with uncertainty.

Meta-active learning optimizes control of safety-critical systems by efficiently learning dynamics and configurations.

A new framework for SPX and VIX hedging that combines AI and market dynamics.

Paper reviews robustness in machine learning models and discusses training and certification methods.

Deep neural network controllers for autonomous driving have recently benefited from significant performance improvements, and have begun deployment in the real world. Prior to their widespread adoption, safety guarantees are needed on the controller behaviour that properly take account of the uncertainty within the mod…

In this work we seek for an approach to integrate safety in the learning process that relies on a partly known state-space model of the system and regards the unknown dynamics as an additive bounded disturbance. We introduce a framework for safely learning a control strategy for a given system with an additive disturba…

Tail-Safe hedging uses reinforcement learning with a safety layer to manage financial risks.

Reinforcement learning is a powerful paradigm for learning optimal policies from experimental data. However, to find optimal policies, most reinforcement learning algorithms explore all possible actions, which may be harmful for real-world systems. As a consequence, learning algorithms are rarely applied on safety-crit…

Gaussian Processes (GPs) are widely employed in control and learning because of their principled treatment of uncertainty. However, tracking uncertainty for iterative, multi-step predictions in general leads to an analytically intractable problem. While approximation methods exist, they do not come with guarantees, mak…

Study improves self-driving safety in dynamic environments.