Log Barriers for Safe Black-box Optimization with Application to Safe Reinforcement Learning

by   Ilnura Usmanova, et al.

Optimizing noisy functions online, when evaluating the objective requires experiments on a deployed system, is a crucial task arising in manufacturing, robotics and many others. Often, constraints on safe inputs are unknown ahead of time, and we only obtain noisy information, indicating how close we are to violating the constraints. Yet, safety must be guaranteed at all times, not only for the final output of the algorithm. We introduce a general approach for seeking a stationary point in high dimensional non-linear stochastic optimization problems in which maintaining safety during learning is crucial. Our approach called LB-SGD is based on applying stochastic gradient descent (SGD) with a carefully chosen adaptive step size to a logarithmic barrier approximation of the original problem. We provide a complete convergence analysis of non-convex, convex, and strongly-convex smooth constrained problems, with first-order and zeroth-order feedback. Our approach yields efficient updates and scales better with dimensionality compared to existing approaches. We empirically compare the sample complexity and the computational cost of our method with existing safe learning approaches. Beyond synthetic benchmarks, we demonstrate the effectiveness of our approach on minimizing constraint violation in policy search tasks in safe reinforcement learning (RL).


Safe Reinforcement Learning via Confidence-Based Filters

Ensuring safety is a crucial challenge when deploying reinforcement lear...

Safe non-smooth black-box optimization with application to policy search

For safety-critical black-box optimization tasks, observations of the co...

Log Barriers for Safe Non-convex Black-box Optimization

We address the problem of minimizing a smooth function f^0(x) over a com...

Safe Policy Optimization with Local Generalized Linear Function Approximations

Safe exploration is a key to applying reinforcement learning (RL) in saf...

Safe Continuous Control with Constrained Model-Based Policy Optimization

The applicability of reinforcement learning (RL) algorithms in real-worl...

Lifting Layers: Analysis and Applications

The great advances of learning-based approaches in image processing and ...

A general sample complexity analysis of vanilla policy gradient

The policy gradient (PG) is one of the most popular methods for solving ...

Please sign up or login with your details

Forgot password? Click here to reset