|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/neuroevolution-genetic-algorithms-and-artificial-neuralnets/
课程评论：没有评论
课程名称：使用原生 JavaScript 的遗传算法与人工神经网络 概述：本短课程将从零开始使用 JavaScript（不使用库）构建一个神经进化模型，该模型能学习玩各种游戏。神经进化是一种强大的机器学习和人工智能方法，利用进化算法来进化神经网络。大多数神经网络使用梯度下降而不是神经进化。然而，Uber 的研究人员在 2017 年指出，简单的结构性神经进化算法与复杂的现代行业标准梯度下降深度学习算法具有竞争力。 深度神经进化：遗传算法是训练深度神经网络以进行强化学习的竞争替代方案。本课程向学生介绍神经进化的原理以及设计和实施神经进化算法所用的技术。课程内容包括以下主题： 1. 神经进化简介：基本原理和应用 2. 进化算法：遗传算法、遗传编程和进化策略 3. 神经网络：类型、架构和训练技术 4. 神经进化算法：应用于神经网络的进化算法 5. 神经进化的应用：游戏和优化问题 6. 高级主题：多目标神经进化、递归神经网络的神经进化和深度神经进化 在这个项目中，我们将遗传进化应用于多个游戏，如自动驾驶汽车、智能帽和“吃豆人”。本课程是我另一个关于从零开始创建人工神经网络的课程的后续课程，在那里我展示了如何在没有库的情况下构建 ANN。在那个项目中，学习过程是通过反向传播（梯度下降）完成的，而本项目使用不同的方法——进化算法。 通过完成本课程，学生将能够深入理解神经进化的原理，并能够为各种应用设计和实施神经进化算法。
In this short course, we will build a Neuroevolution model from scratch using only JavaScript (no libraries) that will learn to play various games.Neuroevolution is a powerful approach to machine learning and artificial intelligence that uses evolutionary algorithms to evolve neural networks. Most neural networks use gradient descent rather than neuroevolution. However, around 2017 researchers at Uber stated they had found that simple structural neuroevolution algorithms were competitive with sophisticated modern industry-standard gradient-descent deep learning algorithms.Deep Neuroevolution: Genetic Algorithms are a Competitive Alternative for Training Deep Neural Networks for Reinforcement LearningThis course introduces students to the principles of neuroevolution and the techniques used to design and implement neuroevolution algorithms.The course covers the following topics:Introduction to neuroevolution: basic principles and applicationsEvolutionary algorithms: genetic algorithms, genetic programming, and evolutionary strategiesNeural networks: types, architectures, and training techniquesNeuroevolution algorithms: evolutionary algorithms applied to neural networksApplications of neuroevolution: games, and optimization problemsAdvanced topics: multi-objective neuroevolution, neuroevolution of recurrent neural networks, and deep neuroevolution.In this project, we have applied GeneticEvolution to multiple games such as self-driving cars, smart caps and flappy bird.This course is a follow-up to my other course about Artificial Neural Networks from scratch, where I show how to create an ANN from scratch without libraries. In that project, the learning process is done using backpropagation(gradient descent), this project uses a different approach. We will use Evolutionary Algorithm.By following this course until the end, students will have a solid understanding of the principles of neuroevolution and the ability to design and implement Neuroevolution algorithms for a variety of applications.
