Regret-Aware Black-Box Optimization with Natural Gradients, Trust-Regions and Entropy Control
Maximilian Hüttenrauch, Gerhard Neumann
- 发表年份
- 2022
- 访问权限
- 开放获取
摘要
Most successful stochastic black-box optimizers, such as CMA-ES, use rankings of the individual samples to obtain a new search distribution. Yet, the use of rankings also introduces several issues such as the underlying optimization objective is often unclear, i.e., we do not optimize the expected fitness. Further, while these algorithms typically produce a high-quality mean estimate of the search distribution, the produced samples can have poor quality as these algorithms are ignorant of the regret. Lastly, noisy fitness function evaluations may result in solutions that are highly sub-optimal on expectation. In contrast, stochastic optimizers that are motivated by policy gradients, such as the Model-based Relative Entropy Stochastic Search (MORE) algorithm, directly optimize the expected fitness function without the use of rankings. MORE can be derived by applying natural policy gradients and compatible function approximation, and is using information theoretic constraints to ensure the stability of the policy update. While MORE does not suffer from the listed limitations, it often cannot achieve state of the art performance in comparison to ranking based methods. We improve MORE by decoupling the update of the mean and covariance of the search distribution allowing for more aggressive updates on the mean while keeping the update on the covariance conservative, an improved entropy scheduling technique based on an evolution path which results in faster convergence and a simplified and more effective model learning approach in comparison to the original paper. We compare our algorithm to state of the art black-box optimization algorithms on standard optimization tasks as well as on episodic RL tasks in robotics where it is also crucial to have small regret. We obtain competitive results on benchmark functions and clearly outperform ranking-based methods in terms of regret on the RL tasks.
关键词
相关论文
一种面向线弧增材制造的电动汽车结构可制造性拓扑优化的双环框架
Qiang Cui, Chuan Yu, Daoqian Yang 等 5 位作者
Robotics and Computer-Integrated Manufacturing · 2026
几何数字孪生:一种用于航空发动机装配精度预测的数字智能模型
Ke Shang, Xin Jin, Teli Xu 等 7 位作者
Robotics and Computer-Integrated Manufacturing · 2026
通过人工智能驱动的机器人技术革新产业
Aryan Chaudhary
Recent Advances in Computer Science and Communications · 2026
新型大口径偏置馈电可展开天线设计与动态性能预测
Chuang Shi, Tianming Liu, Ning Xue 等 9 位作者
Aerospace Science and Technology · 2026