BriefGPT.xyz
May, 2019
基于人口的全局优化方法用于学习RNN中的长期依赖
Population-based Global Optimisation Methods for Learning Long-term Dependencies with RNNs
HTML
PDF
Bryan Lim, Stefan Zohren, Stephen Roberts
TL;DR
本文探讨了使用基于群体的全局优化技术（PBO）训练RNNs以捕获时间序列数据中的长期依赖性，通过在波动率预测应用中测试进化策略（ES）和粒子群优化（PSO），证明了PBO方法通常导致性能改进，而ES在各种体系结构中表现最为稳定。
Abstract
Despite recent innovations in network architectures and loss functions, training
rnns
to learn
long-term dependencies
remains difficult due to challenges with gradient-based optimisation methods. Inspired by the
→