We introduce ES-MAML, a new framework for solving the model agnostic meta learning (MAML) problem based on Evolution Strategies (ES). Existing algorithms for MAML are based on policy gradients, and incur significant difficulties when attempting to estimate second derivatives using backpropagation on stochastic policies. We show how ES can be applied to MAML to obtain an algorithm which avoids the problem of estimating second derivatives, and is also conceptually simple and easy to implement. Moreover, ES-MAML can handle new types of nonsmooth adaptation operators, and other techniques for improving performance and estimation of ES methods become applicable. We show empirically that ES-MAML is competitive with existing methods and often yields better adaptation with fewer queries.

中文翻译：

---

ES-MAML：简单的 Hessian-Free 元学习

我们介绍了 ES-MAML，这是一种基于进化策略 (ES) 解决模型不可知元学习 (MAML) 问题的新框架。现有的 MAML 算法基于策略梯度，并且在尝试使用随机策略的反向传播来估计二阶导数时会遇到很大的困难。我们展示了如何将 ES 应用于 MAML 以获得一种算法，该算法避免了估计二阶导数的问题，并且在概念上也很简单且易于实现。此外，ES-MAML 可以处理新型的非平滑自适应算子，其他用于提高 ES 方法性能和估计的技术变得适用。我们凭经验表明 ES-MAML 与现有方法相比具有竞争力，并且通常会以更少的查询产生更好的适应性。