Natural Language Processing (NLP)

Natural Language Processing (2 Hours)

Module 1Introduction to NLP / Text preprocessing / Tokenisation/ Named Entity Recognition
Module 2Syntax, Grammars and Parsing Techniques
Module 3Features, Augmented Grammars, and Its Application
Module 4Parsing Preference/Deterministic Parsing/Lexicalized and Probabilistic Parsing
Module 5Regular Expression, Morphology and Finite-State Transducers
Module 6N-gram and Part-of-Speech Tagging/Word Sense Disambiguation
Module 7Semantics and Logical Form – Linking Syntax and Semantics/ Local Discourse and Reference
Module 8HMM and Speech Recognition / Applications: Question Answering & Machine Translation

Machine Translation (3 Hours)

Module 1Introduction to Machine Translation
Module 2Rule-based Machine Translation / Corpus-based Machine Translation
Module 3Statistical Machine Translation
Module 4Neural Machine Translation
Module 5Neural Machine Translation with Attention
Module 6Transformer for MT
Module 7Machine Translation Evaluation
Module 8Data Preparation for MT

From Linguistics to Formal Grammars (Automata) (3 Hours)

Module 1Linguistics as Pattern Observation
Module 2Formal Language / Chomsky Hierarchy of Formal Grammars
Module 3Regular Languages
Module 4Context-Free Language
Module 5Mildly Context-Sensitive Languages
Module 6Context-Sensitive
Module 7Recursively Enumerable Language
Module 8Pumping Lemmas

Text Processing & BERT
Question & Ansering Thai BERT (5 Hours)

Module 1NLP Transformer
Module 2Feature Representation / Sequence Models and Seq2Seq Model
Module 3Encoder and Decoder Performance / Inside Encoder and Decoder
Module 4GPT-1/ BERT: Theory / Task Specific Methods / Practical / Fine Tuning
Module 5Other Transformer-based Models and GPT-2
Module 6Introduction to BERT / BERT for QA Model and Huggingface / Tokenization and Pipeline
Module 7Text Classification with Wongnai Dataset
Module 8Text Classification with Custom Dataset / Thai Automatic Question Answering

Web Scraping and Document Classification (3 Hours)

Module 1Overview Web Scraping
Module 2Basic HTML Document and Web Scraping
Module 3Regular Expressions with Python
Module 4Document Classification
Module 5Preprocessing and Problems
Module 6Topic Modeling and Tag Suggestion
Module 7Phishing Web Detection and Sentiment Analysis
Module 8Document Classification Workshop

Image Processing

Introduction and Advanced Image processing (3 Hours)

Module 1Basic image processing Color, Light, and Image Formation
Module 2Convolution / Image Gradient / Corner Detection / SIFT
Module 3Histogram Equalization
Module 4Integral Image
Module 5Local Intensity Distribution Equalization (LIDE)
Module 6Maximum Likelihood
Module 7Mixture Model
Module 8Pixel Sorting

Computer vision (2 Hours)

Module 1What is Computer Vision
Module 2Mathematics for Computer Vision
Module 3Filters, Contrast, Transformation, and Morphology
Module 4Image restoration, Sizing, Noise, Segmentation, and Contours
Module 5Color Detection / Background Subtraction: With Background /
Module 6Adding Noises / Background Subtraction: Without Background
Module 7Line Sensor / Perspective Transform
Module 8Example of Computer Vision

Introduction to Point Cloud (Digital Geometry Processing) (2 Hours)

Module 1Introduction of Point Cloud
Module 2How is Point Cloud Generated
Module 3What Format is Point Cloud Stored
Module 4Point Cloud Processing: Registration and Keypoints
Module 5Feature Descriptors
Module 6Correspondences Estimation
Module 7Transformation Estimation
Module 8Applications and Tools

Introduction to Point Cloud (2D-3D Reconstruction) (3 Hours)

Module 1Point Cloud Registration / 2D to 3D Registration
Module 2Camera and object motion tracking
Module 3Feature Extraction and Matching
Module 42D to 3D Reconstruction and Surface Reconstruction
Module 5Explicit and Implicit Surface
Module 62D to 3D Reconstruction and Vanilla ICP
Module 7Normalization Depth and World Origin
Module 8Tools and Pytorch 3D / 3D Machine Learning, 3D Dataset and ABC / VoxelNet and Geometric Deep learning

Introduction to Point Cloud (3D Deep Learning and PointNet) (2 Hours)

Module 1Introduction 3D Deep Learning and voxel-based
Module 2Point-based and Graph-based
Module 3Point Net – Introduction and Input / Embedding Space, OOS and Result
Module 4Generating Mesh-based Shapes From learned latent of Point Cloud with VAE-GAN
Module 5Point Attention Network for Gesture Recognition Using Point Cloud Data
Module 6Gather, Scatter, and Mesh base / Keypoint 3D and VAE-GAN
Module 7Library Installation and Data Processing / Pytorch-geometric Concept, Data Loader
and Model Function / Create Dataset Using Pytorch
Module 8Model Deployment / Dynamic Graph CNN for Learning on Point Clouds Model / Message Passing Networks

Internet of Things and Robotics

Arm Robotics and Robot Kinematics (6 Hours)

Module 1Introduction to Robotics / The basic principles of robotics
Module 2Robot Types
Module 3Introduction to Robot Mechanics
Module 4Robot Controller
Module 5Sensors and actuators
Module 6Overview Introduction to Robot Kinematics
Module 7The Fundamental Concepts of Kinematics
Module 8Notation and Matrix Representation

Introduction and Advanced Internet of Things (6 Hours)

Module 1Smart IoT Device / IoT Introduction & Computing System / The Internet of Things
Module 2IoT Components / IoT & AIoT Concept / Embedded Board & Programming Languages
Module 3Arduino Introduction / Basic I/O / Simulator – Digital Input-Output / Analog (Input, Output)
Module 4IoT Connectivity / IoT Application
Module 5Tool and Supplements / IoT Network Protocol
Module 6Setup Docker Desktop / Case Study: Temp/Humid Monitoring and Control
Module 7Node-RED Simulation / Node-RED Brain Process
Module 8Databases / Grafana

Data Science

Finance on Reinforcement Learning (2 Hours)

Module 1Introduction Machine Learning in Finance
Module 2Reinforcement Learning (RL) Concept
Module 3Markov Decision Process (MDP), Task, Value Functions, and Optimum Policy
Module 4How to solve RL Problem (Dynamic Programing)
Module 5Q-Learning
Module 6Policy Gradient RL
Module 7Rewards & BackTest
Module 8RL Architecture and Other RL

Stock Manipulation (2 Hours)

Module 1Introduction Recurrent Neural Network (RNN) – Introduction RNN
Module 2RNN, Type of RNN and Applications
Module 3Backpropagation Through Time(BPTT) and Long Short Term Memory(LSTM)
Module 4Activity: understand how RNN works?
Module 5Word2Vec Model, Language Translation and Problem
Module 6Attention Mechanism(Self-Attention)
Module 7Transformer / Detection of Stock Manipulation
Module 8Machine Learning for Manipulation Detection

Signal Processing

Automatic Speech Recognition (ASR) (4 Hours)

Module 1Introduction to Automatic Speech Recognition
Module 2Speech recognizer
Module 3ASR Technique and Partii Demo
Module 4Basic structure ASR System
Module 5KALDI – Introduction
Module 6Example script – common voice
Module 7Data preparation: Thai syllable system
Module 8Language model preparation

Machine Learning

Machine Learning (6 Hours)

Module 1Introduction of Machine Learning
Module 2Pre-machine learning: rule-base/ Supervised Learning/ Unsupervised Learning / Linear Regression: Loss Function
Module 3Evaluation, Labeling, Matrics, and Error / K-means Clustering and KNN Classification
Module 4The centroid for Cosine Similarity / Selecting-K, Variance and Hierarchical Clustering
Module 5Other Flavors of Supervision and Reinforcement Learning / self-supervised
Module 6Model Selection and Literature Review / My Personal Picks (Regression, KNN, XGboost, DL)
Module 7Data size and Class Imbalance
Module 8Bias, Variance and Double Descent Problem / Diagnosis

Advanced Machine Learning (2 Hours)

Module 1SVM and Kernel Methods
Module 2Kernel Function / Dual and Kernel Perceptron
Module 3Margin and Generalization
Module 4Dual Problem
Module 5Maximum Margin Classifier
Module 6Support Vector Machine(SVM)
Module 7Kernel Principal Component Analysis (PCA)
Module 8Fancy Kernel

Ensemble Method (1 Hours)

Module 1Introduction to Ensemble Method
Module 2Bagging Method
Module 3Decision Tree and Random Forest
Module 4Splitting Strategy
Module 5Training Model
Module 6Prediction and Evaluation

One-Class Classification (3 Hours)

Module 1Introduction and Anomaly Detection
Module 2Outlier
Module 3Overview One-Class Classification
Module 4Structure One-Class Classification
Module 5None One-Class Support Vector Machine
Module 6Generative Adversarial Network (GAN)
Module 7Deep One-Class Classification
Module 8Applications: Example 1-4 and Conclusion

Pricipal Component Analysis (PCA) (3 Hours)

Module 1Matrix and Vector
Module 2Subspace Projection and Variance
Module 3Principal Component Analysis (PCA) / Lagrangian
Module 4PCA Axes Selection / PCA Algorithm and Reconstruction Error
Module 5Eigen-Decomposition: Jacobi Method
Module 6PCA: For Small and Large Sample Size Data
Module 7Kernel PCA (KPCA) / Kernel Trick and Summary of KPCA / How to Apply KPCA to Set of 10,000 Vectors
Module 8Related Techniques

Deep Learning

Introduction Deep Learning (5 Hours)

Module 1Introduction to TensorFlow
Module 2Overview of Neural Networks
Module 3Loss Function / Back Propagation / Activation Functions
Module 4Tensor Operations / Linear and Nonlinear Regression
Module 5Deep Learning and MNIST Dataset / One-Hot Encoder and Muti-Layer Perceptron (MLP)
Module 62D Discrete Cross-correlation / Convolutional Neural Network (CNN) / 2D Convolutional layer / The First CNN: LeNet5
Module 7Transfer Learning and Example Transfer Learning
Module 8Deeper Network: VGG, Adam and Example VGG16

Advanced Deep Learning (4 Hours)

Module 1Transfer learning – VGG16
Module 2Autoencoder Feature Extraction and Encoder, Reconstructor and Combine feature
Module 3Autoencoder: Convolutional Neural Network (CNN) UpSampling 2D and Denoising
Module 4Semantic Segmentation (SegNet) Batch Normalization / Lung Sementation (SegNet) U-Net and Dropout / Residual Neural Network (ResNet)
Module 5CIFAR – 10 (Datasets) / Traditional
Module 6Generative Adversarial Networks (GANs) / Super Resolution GAN (SRGAN)
Module 7Audio (STFT, MFCC) Example Audio Recognition (CNN)
Module 8Time series, Simple NN Example Stock Prediction
Module 9Text: Feature Extraction Word2Vec, RNN, and LSTM
Module 10Example: Colab Code Audio, Time serial, and Text

Mathematics for AI

Introduction Mathematics for AI (3 Hours)

Module 1Machine Learning Introduction
Module 2Linear Algebra: Vector and Matrix
Module 3Linear Algebra : Transformation & Determinant
Module 4Linear Algebra: Eigenvectors
Module 5Linear Algebra: Singular Value Decomposition (SVD)
Module 6Probability: Basics of Probability and Weighted and Unweighted
Module 7Calculus: Differential and Integral
Module 8Statistical

Advanced Mathematics for AI (2 Hours)

Module 1Features
Module 2Loss Function
Module 3Binary Cross-Entropy Loss
Module 4Maximum Likelihood
Module 5Linear Classifier
Module 6Multinomial Model
Module 7High Dimensional Data

Programming and Tools

Basic of Python Programming (2 Hours)

Module 1Python Setup / Fundamental Data Types / Comparison and Logical Operators
Module 2Control Flow Statements / IF-Then Statements
Module 3Loops
Module 4Built-in Python Expressions / Functions
Module 5Classes & Objects
Module 6Import Library
Module 7Inputs and Outputs
Module 8Files

Introduction to Linux (2 Hours)

Module 1Linux Structure and Installation
Module 2User Environment/ Text Editors/ Command-line Operations
Module 3Local Security Principles / Network Operations
Module 4Manipulating Text / Bash Shell Scripting / Advanced Bash Shell Scripting
Module 5Processes
Module 6Finding Linux Documentation
Module 7File Operations
Module 8Common Applications

Introduction to Kaggle (30 Minutes)

Module 1Introduction to Kaggle
Module 2Advanced to Kaggle
Module 3Competition to Kaggle
Module 4Kaggle Hands-on

Introduction to Google Colab (1 Hours)

Module 1Introduction to Google Colab
Module 2Google Colab Tour
Module 3Basic Uses
Module 4Advanced Uses and Conclusion
Module 5Google Colab Hands-on

Introduction to GitHub (1 Hours)

Module 1Introduction to GitHub and Version Control (Including Git installation)
Module 2Text Editor and Extension
Module 3Git Using in Project and Basic Commands
Module 4Branch
Module 5Github Registration and Authentication
Module 6Github Repository Management

Microservices and Docker Compose (4 Hours)

Module 1Introduction Microservice / Microservice as RESTful APIs / API Gateway
Module 2Event-driven Communication, Secure Microservices, Data Management, Event Communication
Module 3Success Factor and Grafana
Module 4Introduction to Docker, Build and Run
Module 5Old Style Deploy, The Matrix of Hell, Container Isolation, Glossary
Module 6Install Docker, Command, and Network
Module 7Docker Image / Docker-compose / Services / Networks, Volumes, Docker-compose
Module 8Example and Showcase of Docker-compose

Introduction to Docker (2 Hours)

Module 1Introduction Microservice / Microservice as RESTful APIs / API Gateway
Module 2Event-driven Communication, Secure Microservices, Data Management, Event Communication
Module 3Success Factor and Grafana
Module 4Introduction to Docker, Build and Run
Module 5Old Style Deploy, The Matrix of Hell, Container Isolation, Glossary
Module 6Install Docker, Command, and Network
Module 7Docker Image / Docker-compose / Services / Networks, Volumes, Docker-compose
Module 8Example and Showcase of Docker-compose

Introduction to Deployment (2 Hours)

Module 1Introduction to Application Programming Interface (API) and Deployment
Module 2API Development with Python and Flask
Module 3Machine Learning Model with Python
Module 4Machine learning as a Service Development with API
Module 5Testing API with Postman