Data mining and machine/deep learning for Geophysics

Section outline

• Select section Introduction

  Introduction

  Collapse all Expand all

  ONLINE

  Department of Information and Electrical Engineering and Applied Mathematics, University of Salerno

  

  30 June 2025, from 9:00 to 18:00

  1 July 2025, from 9:00 to 18:00

  2 July 2025, from 9:00 to 18:00

  3 July 2025, from 9:00 to 18:00

   25 (Online)

  

  Training objectives: The course aims to provide the basics of artificial intelligence and the paradigms of machine learning and deep learning.

  The course is organised in four 6-hour modules and the lectures will be theoretical and practical, with thematic discussions and guided exercises.

  Exercises involving the use of Machine Learning frameworks in the Python language for geophysics applications  will be organised as part of the course.

  • Select activity Announcements

    

    Announcements Forum
  • Select activity Description of the course

    

    Description of the course File PDF
• Select section Syllabus

  Syllabus

  • Principles of Artificial Intelligence
  • Machine Learning 1
  • Machine Learning 2
  • Deep Learning
• Select section Principles of Artificial Intelligence

  Principles of Artificial Intelligence

  History and principles of AI, Turing Test, The concept of learning from examples, AI in contemporary society, Cognitive Robotics, Self-driving vehicles, Cybersecurity, Generative AI, AI in medicine.

  • Select activity Modulo 1

    

    Modulo 1 File PDF
• Select section Machine Learning 1

  Machine Learning 1

  Introduction to machine learning. Characteristics and areas of applicability of AI methodologies. Elements of decision theory. Bias and variance error. Handcrafted feature-based approaches. Preprocessing. Curse of dimensionality and dimensionality reduction. Correlation and confounding features. 

  • Select activity Modulo 2

    

    Modulo 2 File PDF
• Select section Machine Learning 2

  Machine Learning 2

  Parametric and non-parametric models, linear models, logistic regression, Support Vector Machines, decision trees and Random Forest. Metrics and protocols for evaluating the performance of classifiers. Cross-validation protocol and selection of hyper-parameters. Effect of regularisation terms in the learning process.

  • Select activity Modulo 3

    

    Modulo 3 File PDF
• Select section Deep Learning

  Deep Learning

  Introduction to deep neural networks: non-linearly separable problems, training from data and representation learning. The multi-level perceptron and the Back-propagation algorithm. Convolutional and recurrent neural networks. The very deep approaches.

  • Select activity Modulo 4

    

    Modulo 4 File PDF