arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1405.4604cs.LGcs.NE
收藏

关于非凸优化的鞍点问题

On the saddle point problem for non-convex optimization

Université de Montréal(蒙特利尔大学);Stanford University(斯坦福大学)

Razvan Pascanu, Yann N. Dauphin, Surya Ganguli, Yoshua Bengio

摘要:

科学和工程的许多领域的一个中心挑战涉及在连续的高维空间上最小化非凸误差函数。梯度下降或拟牛顿方法几乎普遍用于执行这样的最小化,并且通常认为这些局部方法找到全局最小值的能力的困难的主要来源是具有比全局最小值高得多的误差的局部最小值的扩散。在这里,我们认为,统计物理,随机矩阵理论和神经网络理论的结果的基础上,更深层次和更深刻的困难源于鞍点的扩散,而不是局部极小值,特别是在高维问题的实际利益。这样的鞍点被高误差高原所包围,这会大大减缓学习速度,并给人一种存在局部最小值的错觉。受这些论点的启发,我们提出了一种新的算法,鞍无牛顿法,可以快速逃离高维鞍点,不同于梯度下降和拟牛顿法。我们将该算法应用于深度神经网络训练,并为其优越的性能提供了初步的数值证据。

英文摘要:

A central challenge to many fields of science and engineering involves minimizing non-convex error functions over continuous, high dimensional spaces. Gradient descent or quasi-Newton methods are almost ubiquitously used to perform such minimizations, and it is often thought that a main source of difficulty for the ability of these local methods to find the global minimum is the proliferation of local minima with much higher error than the global minimum. Here we argue, based on results from statistical physics, random matrix theory, and neural network theory, that a deeper and more profound difficulty originates from the proliferation of saddle points, not local minima, especially in high dimensional problems of practical interest. Such saddle points are surrounded by high error plateaus that can dramatically slow down learning, and give the illusory impression of the existence of a local minimum. Motivated by these arguments, we propose a new algorithm, the saddle-free Newton method, that can rapidly escape high dimensional saddle points, unlike gradient descent and quasi-Newton methods. We apply this algorithm to deep neural network training, and provide preliminary numerical evidence for its superior performance.

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。

↑