Difference between revisions of "Artificial Neural Networks and Deep Learning"

Revision as of 14:43, 29 November 2019

The following are last minute news you should be aware of ;-)

21/11/2019: From today AN2DL Friday lectures are moved to class 26.16
16/11/2019: Kaggle Homework published together with material on Keras and Tensorflow2 
04/11/2019: Fixed download of first lab material
16/10/2019: New deck of slides uploaded
15/10/2019: Thursday lectures moved to room 2.0.1 indeed!
10/10/2019: Today's lecture moved to Aula Rogers!!!
04/10/2019: Uploded slides about feed forward neural networks
04/10/2019: Today's lecture moved to 4.0.1!!!
26/09/2019: Uploaded slides about deep learning
24/09/2019: Thursday 26/09/2019 lecture will be in room B21 !!!
19/09/2019: No lecture on the 20/09/2019 ... check the detailed schedule.
19/09/2019: The course starts today!

Course Aim & Organization

Neural networks are mature, flexible, and powerful non-linear data-driven models that have successfully been applied to solve complex tasks in science and engineering. The advent of the deep learning paradigm, i.e., the use of (neural) network to simultaneously learn an optimal data representation and the corresponding model, has further boosted neural networks and the data-driven paradigm.

Nowadays, deep neural network can outperform traditional hand-crafted algorithms, achieving human performance in solving many complex tasks, such as natural language processing, text modeling, gene expression modeling, and image recognition. The course provides a broad introduction to neural networks (NN), starting from the traditional feedforward (FFNN) and recurrent (RNN) neural networks, till the most successful deep-learning models such as convolutional neural networks (CNN) and long short-term memories (LSTM).

The course major goal is to provide students with the theoretical background and the practical skills to understand and use NN, and at the same time become familiar and with Deep Learning for solving complex engineering problems.

Teachers

The course is composed of a blending of lectures and exercises by the course teachers and a teaching assistant.

• Matteo Matteucci: the course teacher
• Giacomo Boracchi: the course teacher
• Francesco Lattari: the course teaching assistant

Course Program

This goal is pursued in the course by:

• Presenting major theoretical results underpinning NN (e.g., universal approx, vanishing/exploding gradient, etc.)
• Describing the most important algorithms for NN training (e.g., backpropagation, adaptive gradient algorithms, etc.)
• Illustrating the best practices on how to successfully train and use these models (e.g., dropout, data augmentation, etc.)
• Providing an overview of the most successful Deep Learning architectures (e.g., CNNs, sparse and dense autoencoder, LSTMs for sequence to sequence learning, etc.)
• Providing an overview of the most successful applications with particular emphasis on models for solving visual recognition tasks.

Detailed course schedule

A detailed schedule of the course can be found here; topics are just indicative while days and teachers are correct up to some last minute change (I will notify you by email). Please note that not all days we have lectures!!

Note: Lecture timetable interpretation
* On Thursday, in L26.12, starts at 16:15, ends at 18:15
* On Friday, in 6.0.1, starts at 14:15, ends at 17:15

Course Evaluation

Course evaluation is composed of two parts:

• A written examination covering the whole program graded up to 27/32
• A home project in the form of a Kaggle style competition practicing the topics of the course graded up to 5/32

The final score will sum the grade of the written exam and the grade of the home project.

Teaching Material (the textbook)

Lectures will be based on material from different sources, teachers will provide their slides to students as soon they are available. As a general reference you can check the following text, but keep in mind that teachers will not follow it strictly

• Deep Learning. Ian Goodfellow, Yoshua Bengio, and Aaron Courville, MIT Press, 2016.

Course Slides

Slides from the lectures by Matteo Matteucci

• [2019/2020] Course Introduction: introductory slides of the course with useful information about the course syllabus, grading, and the course logistics.
• [2019/2020] Machine Learning vs Deep Learning: introduction to machine learning paradigms and definition of deep learning with examples
• [2019/2020] From Perceptrons to Feed Forward Neural Networks: the original Perceptron model, Hebbian learning, feed-forward architecture, backpropagation and gradient descent, error functions and maximum likelihood estimation
• [2019/2020] Neural Networks Training: dealing with overfitting (weight decay, early stopping, dropout), vanishing gradient (ReLU and friends), batch normalization
• [2019/2020] Recurrent Neural Networks: learning with sequences, Recurrent Neural Networks, vanishing gradient, Long Short-Term Memories (LSTM).
• [2019/2020] seq2seq:

Slides from the lectures by Giacomo Boracchi are available in his webpage, for you

• Image Classification: Image classification and related issues, template matching, image classification via nearest neighbors methods, image classification via linear classifiers, image classification via hand-crafted features.

Slides from the practicals by Francesco Lattari

• [2019/2020] Introduction to Keras: Introduction to Keras and Tensorflow2 (slides + notebook)
• [2019/2020] Convolutional architectures in Keras: How to build train and evaluation convolutional models for classification and segmenation in Keras and Tensorflow2 (slides + notebook)

Kaggle Homeworks

As part of the evaluation (up to 5 marks in the final grade) we are issuing 3 homeworks in the form of Kaggle competitions. They are meant to practice the course topics on simple image recognition tasks.

• Image Classification Homework: the first homework is about image classification with convolutional neural networks. The deadline to submit the results is November the 30th.
• ...
• ...