Artificial Intelligence - Easily Explained For Beginners

所在平台: Udemy

课程主页: https://www.udemy.com/course/artificial-intelligence-ai-simply-explained-for-beginners/

课程评论:没有评论

第一个写评论        关注课程

课程简介

Coursera 课程《人工智能:极易为初学者讲解》内容概述 本课程旨在为初学者系统性地介绍人工智能(AI)的基础知识,通过梳理AI的历史发展脉络,帮助学习者深入理解AI的核心概念。 课程内容涵盖以下几个主要部分: **第一部分:AI的介绍与历史背景** * **AI的定义**:从哲学角度探讨什么是AI。 * **强AI与弱AI**:区分不同类型的AI。 * **图灵测试**:了解AI领域里程碑式的测试方法。 * **AI的诞生与发展**:回顾AI的初创时期、“伟大期望时代”以及“回归现实”的历程。 * **机器学习入门**:探讨机器如何“学习”的机制。 * **分布式系统在AI中的应用**。 * **核心技术概述**:介绍深度学习、机器学习和自然语言处理等关键技术。 **第二部分:通用问题求解器(General Problem Solver)** * **编程证明**:以“逻辑理论家”(Logical Theorist)为例,介绍早期AI的符号推理能力。 * **“人类问题解决”(Human Problem Solving)中的案例**:以Simon的经典研究为例,展示AI在解决问题结构方面的探索。 * **早期AI技术的概念与代表性系统**:深入了解触发早期AI热潮的理论与实践。 **第三部分:专家系统(Expert Systems)** * **知识表示**:区分事实性知识与启发式知识。 * **知识表示结构**:介绍“框架”(Frames)、“槽”(Slots)和“填充物”(Filler)等概念。 * **推理机制**:讲解“前向推理”(Forward Chaining)和“后向推理”(Backward Chaining)。 * **代表性程序**:以MYCIN程序为例,展示专家系统的构建与应用。 * **专家系统中的概率应用**:通过“发际线裂缝的概率”等案例,说明如何处理不确定性。 * **专家系统特点**:强调其专注于特定领域,并依赖于大量的规则和知识库。 **第四部分:神经网络(Neuronal Networks)** * **生物神经元模型**:模仿人脑神经元的结构与信号处理方式。 * **感知器**:介绍最早的神经网络模型之一。 * **神经网络的回归与挑战**:探讨早期神经网络在复现人脑功能上的尝试,以及帮助其实现突破的关键缺失要素。 **第五部分:机器学习(深度学习与计算机视觉)** * **深度学习的诞生与发展**:追溯深度学习的起源,并介绍其多层网络结构。 * **计算机视觉(Machine Vision / Computer Vision)**:通过“马铃薯收割”等具体案例,展示计算机如何“看”。 * **卷积神经网络(Convolutional Neural Network)**:介绍在图像处理领域取得巨大成功的关键技术。 * **智能体(Agent)与多智能体系统**:探讨智能体概念及其在多智能体系统中的交互,以及如何通过分布式复杂性来解决问题。 **第六部分**:本课程的第六部分(未在概述中编号,但内容已包含在第五部分最后的描述中)着重于多层神经网络、机器学习、计算机视觉、语音识别等现代AI的突破及其应用。 本课程内容系统而全面,适合希望从零开始理解AI理论与发展的初学者。

课程评论(0条)

课程详情

This video course on artificial intelligence is aimed at beginners and is designed to teach you the basics within the historical development of AI. For this reason, our journey begins with the section "Introduction and historical background of AI".Topics and contents of the lessons:I. Introduction and historical backgroundWhat is AI - a philosophical considerationStrong and Weak AIThe Turing TestThe birth of the AIThe era of great expectationsCatching up with realityHow to teach a machine to learnDistributed systems in the AIDeep Learning, Machine Learning, Natural Language ProcessingII. The general problem solverProof Program - Logical TheoristExample from "Human Problem Solving" (Simon)The structure of a problemIn this section, we first take up the initial techniques of AI. You will learn about the concepts and famous example systems that triggered this early phase of euphoria.III. Expert SystemsFactual knowledge and heuristic knowledgeFrames, Slots and FillerForward and backward chainingThe MYCIN ProgrammeProbabilities in expert systemsExample - Probability of hairline cracksIn this section, we discuss expert systems that, similar to the general problem solvers, only deal with specific problems. But instead, they use excessive rules and facts in the form of a knowledge base.IV. Neuronal NetworksThe human neuronSignal processing of a neuronThe PerceptronThis section heralds a return to the idea of being able to reproduce the human brain and thus make it accessible to digital information processing in the form of neural networks. We look at the early approaches and highlight the ideas that were still missing to help neural networks achieve a breakthrough.V. Machine Learning (Deep Learning & Computer Vision)Example - potato harvestThe birth year of Deep LearningLayers of deep learning networksMachine Vision / Computer VisionConvolutional Neural Network.The idea of an agent and its interaction in a multi-agent system is described in the fifth section. The main purpose of such a system is to distribute complexity over several instances.The sixth section deals with the breakthrough of multi-layer neural networks, machine learning, machine vision, speech recognition and some other applications of today's AI.

课程标签

0人关注该课程

主题相关的课程