Part I. Machine Learning for Classification & Clustering

• Introduction to Machine Learning as a Level of Artificial Intelligence

• Design & Implementation of Machine Learning Algorithms

• Machine Learning Types: Clustering Vs Classification Vs Forecasting

• Machine Learning Algorithms (Neural Network, Naive Bayes, SVM, Similarity-based Classification): How to train, validate, and test

• Performance Evaluation of Machine Learning for classification and clustering : Confusion Matrix, Recall, Precision and F1-score

• Practical Session 1: Implementation of Machine Learning for classification

• Practical Session 2: Implementation of Machine Learning for clustering

• Practical Session 3: Deployment of Machine Learning Models on a Web Application

Part II. Machine Learning and Deep Learning for Forecasting

• Forecasting Types: Univariate Vs Multivariate

• Forecasting Terms: Short Vs Mid Vs Long terms

• Deep Learning Architectures Based on Recurrent Neural Networks (RNN)

• A step-by-step explanation of Long Short Term Memory (LSTM): from RNN to LSTM

• Hyperparameters configuration of LSTM for interpretable architectures

• Performance Evaluation of Machine Learning for forecasting: MSE, MAE, and MAPE

• Practical Sessions 4, 5, 6 & 7: Implementation of Machine Learning and deep learning algorithms for Multivariate & Univariate Forecasting

• Practical Session 8: Deployment of Machine and Deep Learning Forecasting Models on a Web Application

Part III. Machine Learning & Deep Learning for Computer Vision: Classification Vs Detection Vs Segmentation

• Why Deep Learning: is crucial for Multimedia Object Representation & Classification?

• Design & Implementation of Deep Learning Architectures

• Deep Learning Architectures based on feed-forward

• Convolutional Neural Network: from scratch to transfer learning

• Using Convolution Neural Network for Detection & Segmentation: Case of Yolo

• Practical Session 9: Implementation of CNN from scratch for Image classification: binary Vs multiclass

• Practical Session 10: Implementation of CNN using transfer learning for Image classification: binary vs. multiclass

• Practical Session 11: Implementation of Yolo for Object Detection & Segmentation

• Practical Session 12: Deployment of Machine and Deep Learning Models on a Web Application for Computer Vision

École d'hiver de la technologie
(21-24 Décembre 2022)

L' école d'hiver de l'embarqué et de l'Intelligence artificielle dans sa quatrième édition a permis au participants de découvrir les différentes plateformes émergentes de nos jours: Raspberry pi 4 et ESP32. Des ateliers diversifiés sur la commande à distance, domotique, robotique, computer vision, IoT, deep learning, machine learning, Natural Language Processing, Datamining, ...