-
Machine Learning
-
Deep Learning
-
AI
0/ From intelligence & learning to AI & ML. The TEFPA framework.
Introduction to list of final projects
1/ Predictions: introduction to linear models
Plot Decision Boundary of Neural Network for Spiral dataset
Spiral Scatter Plot for 3 Classes
Train the Deep Neural Network for Fashion Mnist dataset (epochs = 50)
Train the Neural Network for Spiral dataset
Visualize Fashion Mnist dataset
Evaluate and Inference the visualized Fashion Mnist dataset trained by Deep Neural Network
2/ Predictions: introduction to nonlinear models
3/ Representations: feature extraction, embedding coordinates, and nonlinear transformations.
4/ Evaluation: common metrics and loss functions
5/ Search: gradient descent and variants
6/ More on search: overfitting, underfitting, regularization, and generalization
7/ More on representation: CNNs for grid-like data
8/ More on representation: RNNs for time-series-like data
9/ Decision trees & Ensemble methods in practical use.
10/ Unsupervised learning: Kmeans clustering
11/ Data acquisition, cleaning, annotation.
12/ Midterm exam + Projects open discussion
13/ Sequential decision making: classical MDP planning
14/ Sequential decision making: Tabular Q-learning & DQN
15/Interactive decision making: Contextual & multi-armed bandits
16/ DeepCNN: AlexNet, VGGNet, ResNet, MobileNet, etc.
17/ Computer vision applications: image classification, segmentation, etc.
Review
18/ Theory reviews + Final projects checkpoint + implementation guide
19/ Sequence modeling: LSTM /GRU & language models
20/ NLP applications: sentiment classification, language generation, etc.