Derived from

119411 Informatic Systems in Digital management of agriculture and mountain areas LM69 Ortenzi Luciano
(syllabus)
SUPERVISED LEARNING Introduction, what is machine learning: definitions, concepts and applications, coding (basic knowledge). Linear regression model (Cost function,gradient descent, learning rate, pseudoinverse matrix formula) Multiple features (gradient descent for multiple linear regression) Feature scaling and Z−score,Feature engineering,Polynomial regression Logistic regression, Decision boundary (cost function for logistic regression gradient descent implementation). The problem of overfitting regularization for linear regression and logistic regression
2 UNSUPERVISED LEARNING The clustering problem, the Kmeans algorithm, Optimization objective
kNN algorithm, Anomaly detection algorithm Anomaly detection vs. supervised learning
3 MACHINE LEARNING IN PRACTICE Hyperparameters, and training strategies. Model evaluation model selection, overfitting, underfitting and regularization Baseline level of performance and learning curves Error analysis and iterative loop of ML development Transfer learning: using data from a different task,error metrics for skewed datasets, Trading off precision and recall
4 NEURAL NETWORKS AND DEEP LEARNING
TensorFlow and Matlab implementation Training Details Activation functions (sigmoid, ReLu, etc) Multiclass classification and Softmax and advanced implementations
Advanced Optimization.
Additional Layer Types Convolutional neural network height. Deeplearning applications: Image classification and YOLO
(reference books)
 Abhishek Kumar Pandey, Pramod Singh Rathore, Dr. S. Balamurugan "A Practical Approach for Machine Learning and Deep Learning Algorithms Tools and Technique using MATLAB and Python", BPB Publications, INDIA ISBN: 9789388511131  Aurélien Géron, "HandsOn Machine Learning with ScikitLearn, Keras, and TensorFlow (Concepts, Tools, and Techniques to Build Intelligent Systems)", O'REILLY  Ian Goodfellow_Yoshua Bengio_ Aaron Courville  Deep Learning (2016_ The MIT Press)
