神经网络优化算法与《最优化方法》课程教学融合研究  

Integration of optimization algorithm in neural network model and optimization course teaching

作　　者：孙杰宝[1] 杨畅 吴勃英[1] Sun Jiebao;Yang Chang;Wu Boying(School of Mathematics,Harbin Institute of Technology,Heilongjiang Harbin 150001)

机构地区：[1]哈尔滨工业大学数学学院,黑龙江哈尔滨150001

出　　处：《科技风》2022年第3期113-116,共4页

基　　金：黑龙江省高等教育教学改革研究项目“人工智能驱动下的《最优化方法》课程教学改革”(SJGY20190224);中国高等教育学会理科教育专业委员会高等理科教育研究课题“新时代下数学学科拔尖人才培养模式研究与实践”(20ZSLKJYZD12);哈尔滨工业大学研究生教育教学改革研究重点项目“数学学科杰出人才培养模式探索与实践”。

摘　　要：近年来,人工智能在学术界和工业界都取得了极大的成功,这正是由于深度神经网络在最近十年内的飞速发展。神经网络模型的目标就是最优化,对一个复杂的问题中作出最优决策。几乎所有的神经网络模型都可以建模为一个优化问题。本文就求解神经网络模型过程中涉及的优化算法进行总结,并对《最优化方法》课程教学内容进行了探索。通过将神经网络模型及其优化算法与《最优化方法》课程教学融合,使学生熟悉并掌握其原理和优化求解方法,具备更强的核心竞争力。In recent years,artificial intelligence has achieved great success in academia and industry.This is due to the rapid development of deep neural network in recent ten years.The goal of neural network model is optimization,that is making optimal decision in a complex problem.Almost all neural network models can be modeled as an optimization problem.In this paper,we summary the optimization algorithms involved in the process of solving neural network models,and explore the teaching content of optimization method.Through the integration of neural network model and its optimization algorithm with the teaching of optimization method,students can master its principle and optimization method,and have stronger core competitiveness.

关 键 词：人工智能 最优化方法 神经网络 

分 类 号：G420[文化科学—课程与教学论]