arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1803.01299cs.LG
收藏

深度学习的最优控制方法及其在离散权重神经网络中的应用

An Optimal Control Approach to Deep Learning and Applications to Discrete-Weight Neural Networks

Institute of High Performance Computing(高性能计算研究所)

Qianxiao Li, Shuji Hao

摘要:

深度学习被公式化为离散时间最优控制问题。这使得人们能够表征最优性的必要条件,并开发不依赖于可训练参数的梯度的训练算法。特别是,我们介绍了离散时间逐次逼近法(MSA),这是基于庞特里亚金的最大值原理,用于训练神经网络。得到了离散MSA的严格误差估计,揭示了其动态特性和稳定算法的方法。所开发的方法被施加到训练,在一个相当原则性的方式,神经网络的权重被限制在一个离散集的值。我们获得有竞争力的性能和有趣的是,非常稀疏的权重的情况下,三元网络,这可能是有用的模型部署在低内存设备。

英文摘要:

Deep learning is formulated as a discrete-time optimal control problem. This allows one to characterize necessary conditions for optimality and develop training algorithms that do not rely on gradients with respect to the trainable parameters. In particular, we introduce the discrete-time method of successive approximations (MSA), which is based on the Pontryagin's maximum principle, for training neural networks. A rigorous error estimate for the discrete MSA is obtained, which sheds light on its dynamics and the means to stabilize the algorithm. The developed methods are applied to train, in a rather principled way, neural networks with weights that are constrained to take values in a discrete set. We obtain competitive performance and interestingly, very sparse weights in the case of ternary networks, which may be useful in model deployment in low-memory devices.

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。

↑