• You can log in to your eeworld account to continue watching:
  • (3-3)Gradient Descent (Demo by Minecraft)
  • Login
  • Duration:1 minutes and 40 seconds
  • Date:2019/03/22
  • Uploader:木犯001号
Introduction

The machine learning course taught by the well-received teacher Li Hongyi of National Taiwan University is also one of the very rare courses of this kind to be taught in Chinese. Friends who are not good at foreign languages ​​are in luck. The course focuses on deep learning and emphasizes practicality. In addition to the explanation of basic knowledge and algorithms, it also includes the interpretation of various related cutting-edge technologies. The projects involved in the course are very new and keep pace with the times.

Teacher Li's lectures focus on explaining the basic principles in a simple and easy-to-understand manner. He is humorous and easy to understand. He often uses animation images to explain the principles by analogy. It is really a breath of fresh air in machine learning tutorials~

The reprinted video is for communication and learning only. The original address of the reprint is: https://www.youtube.com/watch?v=CXgbekl66jc&list=PLJV_el3uVTsPy9oCRY30oBPNLCo89yu49
Unfold ↓

You Might Like

Recommended Posts

Regarding the directory structure of txt files in the root directory
I want to use a single-chip microcomputer to create a txt document directly in the root directory of a USB disk through a USB interface chip. The information says that a txt document occupies a 32-byt
qiyelian Embedded System
eZ430-chronos is out of power
[size=4]When I looked at the time today, I felt something was a little strange. It was flashing very fast and the color was a little white. I looked again and saw nothing but a battery mark. It seemed
johnrey TI Technology Forum
Program FAQ
uchar shuzi[] = {"0123456789."}; //Define array uchar tishi[] = {"The volt is:"}; //Define array DispNChar(2,0,12,tishi); //Display prompt information 12 should be the array length 2, what are 0 respe
不随风 Microcontroller MCU
[TI DLP Creative Collection] + Mobile Portable Projection Devices
This can satisfy our need to display pictures, documents, etc. anytime and anywhere, and it is a very good helper for our expression and demonstration. 1. First of all, it must be small, preferably ab
chenzhufly TI Technology Forum
Is there anyone who uses Freescale DSP MSC8157?
Is there anyone who has used Freescale DSP MSC8157? There is too little information about this. Is there any company that can provide training?
gulfweed NXP MCU
MCU ultra-low power consumption ULPMark-CP benchmark ranking
The following figure shows the top 20 rankings of MCU ultra-low power consumption ULPMark-CP. The higher the score, the better.It can be seen that all of the MCUs are based on the ARM Cortex-M core.Wh
火辣西米秀 Microcontroller MCU

推荐文章

连接性使汽车定位更加精准 2025年01月15日
尽管 CDNN 工具包等解决方案允许神经网络在嵌入式处理器上本地运行,但最有效的 AI 系统需要云连接。当连接到数据中心时,嵌入式设备在推理过程(目标应用人工智能模型概述的逻辑做出决策的过程)期间捕获的新信息或异常信息可以反馈到人工智能模型中,以随着时间的推移对其进行优化。 除此之外,ADAS 和自动驾驶用例中连接的主要应用是定位。例如,精密全球导航卫星系统 (GNSS)...
STM32Cube.AI将神经网络转换为STM32的优化代码 2024年08月08日
STM32Cube.AI是业界最先进的工具包,能够与流行的深度学习库互操作,以转换任何用于 STM32 微控制器的人工神经网络( MCU)运行优化推理。该软件套件包括 X-Cube 扩展软件X-CUBE-AI 、为我们的SensorTile 开发套件 (STEVAL-STLKT01V1)提供应用示例的功能包FP-AI-SENSING1,以及对我们的ST BLE 传感器的更新为...
人工神经网络(Artificial Neural Network)算法简介 2024年07月18日
人工神经网络,简称神经网络,是一种模仿生物神经网络的结构和功能的数学模型或者计算模型。其实是一种与贝叶斯网络很像的一种算法。之前看过一些内容始终云里雾里,这次决定写一篇博客。弄懂这个基本原理,毕竟现在深度学习太火了。 神经网络是一种方法,既可以用来做有监督的任务,如分类、视觉识别等,也可以用作无监督的任务。首先,我们看一个简单的例子。如下图所示(这个图网上有很多人引用了,...
什么是神经网络?为什么说神经网络很重要?神经网络如何工作? 2024年07月16日
什么是神经网络? 神经网络是一个具有相连节点层的计算模型,其分层结构与大脑中的神经元网络结构相似。神经网络可通过数据进行学习,因此,可训练其识别模式、对数据分类和预测未来事件。 神经网络将您的输入细分为多个抽象层。比如,可通过大量示例训练其识别模式为语音还是图像,就像人类大脑的行为一样。神经网络的行为由其各个元素的连接方式以及这些连接的强度或权重确定。在训练期间,系统会根据指...

Recommended Content

Circuit

可能感兴趣器件

EEWorld
subscription
account

EEWorld
service
account

Automotive
development
circle

About Us Customer Service Contact Information Datasheet Sitemap LatestNews


Room 1530, 15th Floor, Building B, No.18 Zhongguancun Street, Haidian District, Beijing, Postal Code: 100190 China Telephone: 008610 8235 0740

Copyright © 2005-2024 EEWORLD.com.cn, Inc. All rights reserved 京B2-20211791 京ICP备10001474号-1 电信业务审批[2006]字第258号函 京公网安备 11010802033920号