Introduction to Neural Networks
Informacje ogólne
Kod przedmiotu: | 1100-INN |
Kod Erasmus / ISCED: | (brak danych) / (brak danych) |
Nazwa przedmiotu: | Introduction to Neural Networks |
Jednostka: | Wydział Fizyki |
Grupy: |
Fizyka, II stopień; przedmioty sp. Matematyczne i komputerowe modelowanie procesów fizycznych Fizyka, II stopień; przedmioty z zakresu analizy numerycznej Physics (Studies in English), 2nd cycle; courses from list "Numerical Analysis" Physics (Studies in English); 2nd cycle |
Punkty ECTS i inne: |
(brak)
|
Język prowadzenia: | angielski |
Założenia (opisowo): | Basic Knowledge of python is required. |
Skrócony opis: |
Neural networks are a beautiful biologically-inspired programming paradigm that enables a computer to learn from observational data. In this course, we will learn many of the core concepts behind neural networks. We will start the journey from the origin of neural networks to the modern days. |
Pełny opis: |
Neural networks are a beautiful biologically-inspired programming paradigm that enables a computer to learn from observational data. In this course, we will learn many of the core concepts behind neural networks. We will start the journey from the origin of neural networks to the modern days. Basic Knowledge of python is required. 1-Introduction to neural networks. History and Inspiration from neuroscience. 2-Threshold Logic Units 3-The Perceptron 4-The Hopefield Model 5-Optimization problems 6-Linear Regression 7-Feature Learning 8-Fully Connected Neural Networks 9-Backpropagation 10- Introduction to Keras and Tensorflow 11-Image Classification 12-Sequences, Time Series, and Prediction |
Metody i kryteria oceniania: |
(1) Required to complete a number of computer-based projects in the lab explaining the solutions, (2) The students are asked to prepare a short presentation summarizing one topic from the lectures. |
Właścicielem praw autorskich jest Uniwersytet Warszawski, Wydział Fizyki.