arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1604.08459math.OC
收藏

噪声Optimization:基于比较算法的快速收敛速度

Noisy Optimization: Fast Convergence Rates with Comparison-Based Algorithms

Marie-Liesse Cauwet, Olivier Teytaud

摘要:

无导数优化算法是解决黑箱最优化问题的一种有效方法。当涉及到噪声函数时,经典的基于比较的算法比基于梯度的算法慢。对于二次函数,当$N$是函数评估的数量时,没有大突变的进化算法最多有一个简单的遗憾$O(1/ \sqrt{N})$,而随机梯度下降可以在$O(1/N)$中(紧密地)达到一个简单的遗憾。已经证明,通过有限差分的梯度近似(因此,不是基于比较的方法)对于达到这样的$O(1/N)$是必要的。我们回答这个猜想的否定,提供一个基于比较的算法一样好的梯度法,即达到$O(1/N)$ -的条件下,但是,噪声是高斯。实验结果证实了$O(1/N)$简单的遗憾,即,与$O(1/\sqrt{N})$上发表的许多结果相比,

英文摘要:

Derivative Free Optimization is known to be an efficient and robust method to tackle the black-box optimization problem. When it comes to noisy functions, classical comparison-based algorithms are slower than gradient-based algorithms. For quadratic functions, Evolutionary Algorithms without large mutations have a simple regret at best $O(1/ \sqrt{N})$ when $N$ is the number of function evaluations, whereas stochastic gradient descent can reach (tightly) a simple regret in $O(1/N)$. It has been conjectured that gradient approximation by finite differences (hence, not a comparison-based method) is necessary for reaching such a $O(1/N)$. We answer this conjecture in the negative, providing a comparison-based algorithm as good as gradient methods, i.e. reaching $O(1/N)$ - under the condition, however, that the noise is Gaussian. Experimental results confirm the $O(1/N)$ simple regret, i.e., squared rate compared to many published results at $O(1/\sqrt{N})$.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑