Deep Learning Theory plus Practical Project Course
Content:
Theoretical part: Introduction to machine learning, introduction to classical models of deep learning
(Multilayer Perceptron, Convolutional Neural Network, Recurrent Neural Network, Attention, Encoder/Decoder, Transformer),
and also select some good articles to read intensively.
Practical part: Mainly focus on the combination of natural vector theory and deep learning in Yau's biomathematics team.
Through our research team's real-time research ideas, we can show students how to use deep learning to do research projects.
You are also welcome to bring your project to the lecture and do it as you go!
Characteristic:
It is a theoretical plus practical innovation course
learning by doing
doing by learning
The brain is open, and the ideas are wonderful
Theoretical part: Introduction to machine learning, introduction to classical models of deep learning
(Multilayer Perceptron, Convolutional Neural Network, Recurrent Neural Network, Attention, Encoder/Decoder, Transformer),
and also select some good articles to read intensively.
Practical part: Mainly focus on the combination of natural vector theory and deep learning in Yau's biomathematics team.
Through our research team's real-time research ideas, we can show students how to use deep learning to do research projects.
You are also welcome to bring your project to the lecture and do it as you go!
Characteristic:
It is a theoretical plus practical innovation course
learning by doing
doing by learning
The brain is open, and the ideas are wonderful
讲师
日期
2024年09月10日 至 12月10日
位置
| Weekday | Time | Venue | Online | ID | Password |
|---|---|---|---|---|---|
| 周二,周四 | 09:50 - 11:25 | A3-2a-201 | ZOOM 12 | 815 762 8413 | BIMSA |
修课要求
Algebra, Calculus, Probability, Statistics, Python, Spring Deep Learning course
课程大纲
Syllabus:
Lecture 1: Introduction
Lecture 2: Overview, AI early stage development, etc.
Lecture 3: Jupyter Python install, Deep Learning introduction
Lecture 4: Linear Neural Networks for Regression and Linear Neural Networks for Classification
Lecture 5: Project 1: 阿茨海默症数据集Alzheimer’s disease cell-type-specific gene expression dataset analysis
Lecture 6: UMAP, t-SNE, Multilayer Perceptrons (MP)
Lecture 7: Convolutional Neural Networks (CNN)
Lecture 8: Modern Convolutional Neural Networks (AlexNet, VGG, NiN, GoogLeNet, ResNet, DenseNet)
Lecture 9: Project 2: 优化问题的深度学习求解
Lecture 10: Project 3: 人脸分类
Lecture 11: Recurrent Neural Networks (RNN)
Lecture 12: PCA, SHAP
Lecture 13: Attention Mechanisms and Transformers
Lecture 14: Project 4: 基因预测
Lecture 15: Natural Language Processing: Pretraining (NLP)
Lecture 16: Natural Language Processing: Applications (NLP)
Lecture 17: xBost,RFE, MI
Lecture 18: Project 5: 使用预训练框架写作中文诗歌
Lecture 19: Project 6: 生成图像
Lecture 20: Project 7:
Lecture 21: Project 8:
Lecture 22: Project 9:
Lecture 23: Project 10:
Lecture 24: 结束语
深度学习理论加实践项目课程
内容：
理论部分：机器学习概论，深度学习经典模型介绍（多层感知机，卷积神经网络，循环神经网络，Attention, Encoder/Decoder, Transformer),也选一些好文章来精读。
实践部分：重点介绍丘成栋生物数学团队的自然向量理论和深度学习相结合。通过我们科研团队实时研究思路，设想到成果来向学生展示如何用深度学习做科研项目。也欢迎大家带着项目来听课，边听边做！
特色:
是一门理论加实践创新课程
边学边干
边干边学
脑洞打开，奇思妙想
Lecture 1: Introduction
Lecture 2: Overview, AI early stage development, etc.
Lecture 3: Jupyter Python install, Deep Learning introduction
Lecture 4: Linear Neural Networks for Regression and Linear Neural Networks for Classification
Lecture 5: Project 1: 阿茨海默症数据集Alzheimer’s disease cell-type-specific gene expression dataset analysis
Lecture 6: UMAP, t-SNE, Multilayer Perceptrons (MP)
Lecture 7: Convolutional Neural Networks (CNN)
Lecture 8: Modern Convolutional Neural Networks (AlexNet, VGG, NiN, GoogLeNet, ResNet, DenseNet)
Lecture 9: Project 2: 优化问题的深度学习求解
Lecture 10: Project 3: 人脸分类
Lecture 11: Recurrent Neural Networks (RNN)
Lecture 12: PCA, SHAP
Lecture 13: Attention Mechanisms and Transformers
Lecture 14: Project 4: 基因预测
Lecture 15: Natural Language Processing: Pretraining (NLP)
Lecture 16: Natural Language Processing: Applications (NLP)
Lecture 17: xBost,RFE, MI
Lecture 18: Project 5: 使用预训练框架写作中文诗歌
Lecture 19: Project 6: 生成图像
Lecture 20: Project 7:
Lecture 21: Project 8:
Lecture 22: Project 9:
Lecture 23: Project 10:
Lecture 24: 结束语
深度学习理论加实践项目课程
内容：
理论部分：机器学习概论，深度学习经典模型介绍（多层感知机，卷积神经网络，循环神经网络，Attention, Encoder/Decoder, Transformer),也选一些好文章来精读。
实践部分：重点介绍丘成栋生物数学团队的自然向量理论和深度学习相结合。通过我们科研团队实时研究思路，设想到成果来向学生展示如何用深度学习做科研项目。也欢迎大家带着项目来听课，边听边做！
特色:
是一门理论加实践创新课程
边学边干
边干边学
脑洞打开，奇思妙想
参考资料
https://bimsa.net/activity/DeeLea/
Material:
English Version https://d2l.ai/
Chinese Version https://zh.d2l.ai/
https://developers.google.com/machine-learning/
Material:
English Version https://d2l.ai/
Chinese Version https://zh.d2l.ai/
https://developers.google.com/machine-learning/
听众
Graduate
, 博士后
, Researcher
视频公开
公开
笔记公开
公开
语言
中文
, 英文
讲师介绍
博士毕业后主要从事无线通信领域方面的工作，先后在朗讯，阿尔卡特-朗讯，诺基亚公司任职，资深工程师，具有23年无线通讯领域的丰富知识和经验。目前就职于北京雁栖湖应用数学研究院的研究员从事大数据，区块链，人工智能和机器学习方面的研究。