MATLAB is an interactive programming software developed by MathWorks in the USA, used for scientific computing and engineering simulation. It integrates a wide range of powerful functions, including symbolic computation, numerical analysis, matrix computation, scientific data visualization, data processing, machine learning, image processing, signal processing, computational finance, computational biology, and modeling and simulation of nonlinear dynamic systems, all within an easy-to-use window environment, providing a comprehensive solution for scientific research and engineering design across various scientific fields that require effective numerical computations.
To further promote the research work in universities, research institutes, and enterprises, we invite professors from Tsinghua University to jointly hold the “MATLAB Data Analysis, Image Processing, and Machine Learning” training course. We have already conducted 18 training sessions in Beijing, with over a hundred enterprises, universities, and research institutes participating, totaling hundreds of trained students. The training has significantly improved students’ research capabilities in MATLAB data analysis, image processing, Simulink simulation, algorithm development, machine learning, and deep learning, enabling them to solve practical research problems. Through hands-on practice and training, students have mastered the methods of using MATLAB software and received unanimous praise.
1. Through course study, understand and master the programming syntax of MATLAB software, the use of toolboxes, and various plotting techniques, including 2D plots, 3D plots, terrain maps, interactive editing graphics, and animation drawing, while explaining scientific computing and its visualization through examples;
2. Learn to use common analytical tools to analyze data, providing more reliable data analysis capabilities for scientific research;
3. Explain MATLAB optimization modeling and solving, Simulink modeling and simulation in conjunction with engineering application examples;
4. Master application skills and detailed analysis of MATLAB algorithm development, image processing, machine learning, and deep learning;
5. Be able to choose appropriate algorithm models based on data distribution and write code, using MATLAB software to solve practical application projects and research problems.
1. Participate in online training, and I can attend the same online and offline courses for free afterwards, without limit on the number of times, until mastered.
2. After registration and payment, obtain electronic handouts, previous videos, and models in advance for pre-study;
3. The trainers have rich theoretical and engineering experience, and we will prepare lessons based on students’ actual needs and supplement relevant content;
4. With years of online training experience, we can ensure training quality, recorded videos can be watched unlimited times;
5. After the training ends, trainers will provide students with their phone numbers and emails for technical support, ensuring effective outcomes after training;
6.This course is limited to 40 participants, so please register quickly.The first 20 registrants can receive training videos and materials from previous sessions.
October 25, 2024 — October 27, 2024
Beijing on-site/Tencent Meeting simultaneous teaching for 3 days
-
Using the Tencent Meeting platform for teaching, with years of online training experience to ensure training quality. Recorded videos can be watched unlimited times;
-
After registration and payment, obtain electronic handouts and models in advance for pre-study;
-
After the training ends, trainers will provide students with their phone numbers and emails for technical support, ensuring effective outcomes after training.
Senior experts from research institutions such as the Chinese Academy of Sciences and Tsinghua University. They are mainly engaged in research work in machine learning and data mining, data visualization and software development, system modeling and simulation, with rich research experience, proficient in tools such as MATLAB, Python, deep learning, PyTorch, TensorFlow, Keras, neural networks, support vector machines, decision trees, random forests, as well as swarm optimization algorithms such as genetic algorithms, ant colony algorithms, bat algorithms, etc. In recent years, they have been conducting research on core technologies of deep learning, leading and participating in several key project developments and funding projects, and have authored books such as “30 Case Analyses of MATLAB Intelligent Algorithms” and “43 Case Analyses of MATLAB Neural Networks”. They have published multiple high-level international academic research papers.
|
Course Chapters |
Main Content |
|
Chapter 1 MATLAB Basic Programming Overview |
1. Basic operations in MATLAB: including matrix operations, logical and flow control, functions and scripts, basic plotting, etc. 2. File import: formats such as mat, txt, xls, csv, jpg, wav, avi, etc. 3. MATLAB programming habits, styles, and debugging skills. 4. Vectorized programming and memory optimization. 5. Introduction to digital image processing in MATLAB (common image formats and reading/writing, image type conversion, basic operations on digital images, geometric transformations of digital images, image denoising and restoration, edge detection and image segmentation). 6. Case Study:Heart rate calculation based on smartphone camera 7. Hands-on practice |
|
Chapter 2 Introduction to New Features of MATLAB 2023a |
1. Introduction and demonstration of Live Script and Control functionalities. 2. Introduction and demonstration of batch large data import and Datastore class functions. 3. Introduction and demonstration of Data Cleaning functionality. 4. Introduction and demonstration of Experiment Manager functionality. 5. Overview of MATLAB Deep Learning Toolbox. 6. Introduction to MATLAB Deep Learning Model Hub. 7. Introduction and demonstration of Deep Network Designer functionality. 8. Introduction and demonstration of collaborative functionalities between MATLAB and deep learning frameworks like TensorFlow and PyTorch. |
|
Chapter 3 BP Neural Network |
1. Basic concepts of artificial intelligence (distinguishing regression fitting problems from classification recognition problems; supervised (teacher-guided) learning vs. unsupervised (non-teacher) learning; training set, validation set, and test set; overfitting vs. underfitting). 2. Working principle of BP neural network. 3. Data preprocessing (normalization, outlier removal, data augmentation techniques, etc.). 4. Cross-validation and model parameter optimization. 5. Model evaluation and selection of metrics (regression fitting problems vs. classification recognition problems). 6. Case Study: 1) Handwritten digit recognition 2) Face orientation recognition 3) Regression fitting prediction 7. Hands-on practice |
|
Chapter 4 Support Vector Machine, Decision Tree, and Random Forest |
1. Basic principles of support vector machines (the essence of support vectors, the significance of kernel functions). 2. Basic principles of decision trees (information entropy and information gain; differences between ID3 and C4.5). 3. Basic principles of random forests (why we need random forest algorithms? What do the broad and narrow meanings of “random forest” refer to? Where does the “randomness” manifest?). 4. Knowledge extension: Besides building models, what else can support vector machines and decision trees help us with? How to interpret the results of random forests? 5. Case Study: 1) Iris flower classification (SVM, decision tree) 2) Intelligent diagnosis model for benign/malignant breast cancer based on random forests 6. Hands-on practice |
|
Chapter 5 Dimensionality Reduction and Feature Selection |
1. Dimensionality reduction (Dimension reduction) vs. feature selection (Feature selection), their conceptual differences and relationships. 2. Basic principles of Principal Component Analysis (PCA). 3. Basic principles of Partial Least Squares (PLS). 4. Code implementation of PCA and PLS. 5. Insights from PCA: Judging the rationality of training set and test set division. 6. Classic feature selection methods. (1) Forward selection and backward elimination methods. (2) No-information variable elimination method. (3) Feature selection based on binary genetic algorithms. |
|
Chapter 6 Convolutional Neural Networks |
1. Differences and relationships between deep learning and traditional machine learning (Is having more hidden layers in neural networks always better? What is the essential difference between deep learning and traditional machine learning?). 2. Basic principles of convolutional neural networks (What is a convolution kernel? What is the typical topology structure of CNN? What is the weight-sharing mechanism of CNN? What features does CNN extract?). 3. Differences and relationships among classic deep neural networks such as LeNet, AlexNet, Vgg-16/19, GoogLeNet, ResNet. 4. Downloading and installing pre-trained models (Alexnet, Vgg-16/19, GoogLeNet, ResNet). 5. Case Study: 1) Implementing object recognition using pre-trained CNN models. 2) Using convolutional neural networks to extract abstract features. 3) Customizing the topology structure of convolutional neural networks. 4) 1D CNN model solving regression fitting prediction problems. 6. Hands-on practice |
|
Chapter 7 Network Optimization and Tuning Techniques |
1. Optimizing network topology structure. 2. Optimization algorithms (gradient descent, stochastic gradient descent, mini-batch stochastic gradient descent, momentum method, Adam, etc.). 3. Tuning techniques (parameter initialization, data preprocessing, data augmentation, batch normalization, hyperparameter optimization, network regularization, etc.). 4. Case Study:Optimizing convolutional neural network models. 5. Hands-on practice |
|
Chapter 8 Transfer Learning Algorithms |
1. Basic principles of transfer learning algorithms (Why do we need transfer learning? Why can we transfer learning? What is the basic idea of transfer learning?). 2. Transfer learning algorithms based on deep neural network models. 3. Case Study:Dogs vs. Cats 4. Hands-on practice |
|
Chapter 9 Generative Adversarial Networks (GAN) |
1. What are generative adversarial networks? Why do we need them? What can they help us with? Insights from GAN. 2. Basic principles of GAN and its evolution history. 3. Case Study:Implementing GAN in MATLAB (automatic generation of sunflower images). 4. Hands-on practice |
|
Chapter 10 Recurrent Neural Networks and Long Short-Term Memory Networks |
1. Basic principles of Recurrent Neural Networks (RNN). 2. Basic principles of Long Short-Term Memory (LSTM) networks. 3. Differences and relationships between RNN and LSTM. 4. Case Study: (1) Time series prediction. (2) Sequence-to-sequence classification. 5. Hands-on practice |
|
Chapter 11 Deep Learning-based Video Classification Practical Case |
1. Basic principles of deep learning-based video classification. 2. Reading video stream files and extracting image frames. 3. Using pre-trained CNN models to extract feature maps from specified layers. 4. Customizing and building LSTM neural network models. 5. Case Study:Video classification using the HMDB51 dataset. 6. Hands-on practice |
|
Chapter 12 Object Detection using YOLO Model |
1. What is object detection? Differences and relationships between object detection and recognition. 2. Working principles of the YOLO model. 3. The evolution of YOLO from v1 to v5. 4. Case Study: (1) Implementing real-time object detection on images and videos using pre-trained models. (2) Training your own dataset: mask recognition during the COVID-19 pandemic. 5. Hands-on practice |
|
Chapter 13 U-Net Model |
1. Introduction to semantic segmentation. 2. Basic principles of the U-Net model. 3. Case Study:Semantic segmentation of multispectral images based on U-Net. 4. Hands-on practice |
|
Chapter 14 Autoencoders |
1. Composition and basic working principles of autoencoders. 2. Variants of autoencoders (stacked autoencoders, sparse autoencoders, denoising autoencoders, convolutional autoencoders, masked autoencoders, etc.) and their working principles. 3. Case Study:Image classification based on autoencoders. 4. Hands-on practice |
|
Chapter 15 Discussion and Q&A |
1. How to search for literature? (Do you know how to use Google Scholar, Sci-Hub, ResearchGate? Where to find data and code that accompany papers?). 2. How to refine and discover innovative points? (If it is difficult to make original contributions at the algorithm level, how to combine your actual problems to refine and discover innovative points?). 3. Sharing and copying related learning materials (book recommendations, online course recommendations, etc.). 4. Establishing a WeChat group for future discussions and Q&A. |
There are 3 categories of training fees, please choose flexibly according to your needs.
Category A: Fee is 3900 yuan/person (including training fee, materials fee, A category certificate fee, guidance fee, invoice fee, etc.), accommodation is self-catered.
Certificate: A course completion certificate issued by the Zhongke Software Research Center (Beijing);
Category B: Fee is 4800 yuan/person (including training fee, materials fee, B category certificate fee, guidance fee, invoice fee, etc.), accommodation is self-catered.
Certificate: A senior “MATLAB Software Application Engineer” professional technical talent skill certificate issued by the Career Development Planning Committee of the China Smart Engineering Research Association, included in the committee database, nationwide verification available, can serve as valid proof for promotion and rating.
Category C: Fee is 5800 yuan/person (including conference fee, materials fee, B category + C category certificate fee, guidance fee, invoice fee, etc.), accommodation is self-catered.
Certificate: Passing the exam can obtain the “Senior Artificial Intelligence Application Engineer” vocational technical certificate issued by the Education and Examination Center of the Ministry of Industry and Information Technology of the People’s Republic of China, which can serve as proof of professional technical personnel’s vocational ability assessment, and an important basis for employment, appointment, grading, and promotion of professional technical personnel, nationally recognized and verifiable on the official website.
Note: Conference notifications can be provided, fees can be charged and invoices issued, invoices can be issued in advance, and public transfers can be made afterwards; invoices can be issued for training fees, conference fees, conference registration fees, materials fees, technical service fees, testing fees, etc. The travel expenses and accommodation fees for this offline conference are self-catered.
1. Students can get a discount of 300 yuan with a student ID;
2. Groups of 2 or more (including) can reduce 200 yuan per person;
3. Groups of 3 or more (including) can reduce 300 yuan per person;
4. Groups of 4 or more (including) can reduce 400 yuan per person;
5. Groups of 5 or more (including) will receive one additional free spot;
6. The above discount policies cannot be enjoyed simultaneously, only one can be chosen.
Please scan the QR code below to register online. After successful registration, we will send you the training notification and confirm by phone.
