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

From Chrome
Jump to: navigation, search
(Teaching Material (the textbook))
Line 127: Line 127:
 
*[[Media:AN2DL_02_2019_Prceptron_2_FeedForward.pdf|[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   
 
*[[Media:AN2DL_02_2019_Prceptron_2_FeedForward.pdf|[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   
  
Slides from the lectures by Giacomo Boracchi
+
Slides from the lectures by Giacomo Boracchi are avaible in [http://home.deib.polimi.it/boracchi/teaching/AN2DL.htm his webpage], for you
* ... coming soon ...
+
* [http://home.deib.polimi.it/boracchi/teaching/AN2DL/2019_AN2DL_Lez1_ImageClassification.pdf 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.  
  
 
Additional material from the teachers
 
Additional material from the teachers
 
* ... coming soon ...
 
* ... coming soon ...

Revision as of 14:34, 13 October 2019


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

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.

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
Date Day Time Room Teacher Type Topic
19/09/2019 Thursday 16:15 - 17:15 L26.12 Lecture Giacomo Boracchi Course Introduction
20/09/2019 Friday 14:15 - 17:15 6.0.1 -- -- -- No Lecture Today --
26/09/2019 Thursday 16:15 - 18:15 B.2.1 Lecture Matteo Matteucci Introduction to Machine Learning
27/09/2019 Friday 14:15 - 17:15 6.0.1 Lecture Matteo Matteucci Perceptron and Hebian Learning
03/10/2019 Thursday 16:15 - 18:15 L26.12 -- -- -- No Lecture Today --
04/10/2019 Friday 14:15 - 17:15 4.0.1 Lecture Matteo Matteucci Feedforward neural networks and Backpropagation
10/10/2019 Thursday 16:15 - 18:15 Aula Rogrrs Lecture Matteo Matteucci Error Functions
11/10/2019 Friday 14:15 - 16:15 6.0.1 Lecture Giacomo Boracchi Introduction to Image Classification
17/10/2019 Thursday 16:15 - 18:15 L26.12 Lecture Matteo Matteucci Error functions
18/10/2019 Friday 14:15 - 17:15 6.0.1 Lecture Matteo Matteucci Facing Overfitting
24/10/2019 Thursday 16:15 - 18:15 L26.12 Lecture Matteo Matteucci Training tricks and Vanishing gradient
25/10/2019 Friday 14:15 - 17:15 6.0.1 Practicals Francesco Lattari Keras examples ...
31/10/2019 Thursday 16:15 - 18:15 L26.12 Lecture Giacomo Boracchi Introduction to Convolutional Neural Networks
01/11/2019 Friday 14:15 - 17:15 6.0.1 -- -- -- No Lecture Today --
07/11/2019 Thursday 16:15 - 18:15 L26.12 Lecture Giacomo Boracchi CNN architectures
08/11/2019 Friday 14:15 - 16:15 6.0.1 Lecture Giacomo Boracchi Training with data scarcity
14/11/2019 Thursday 16:15 - 18:15 L26.12 Lecture Giacomo Boracchi CNN for image segmentation
15/11/2019 Friday 14:15 - 17:15 6.0.1 Practicals Francesco Lattari Keras examples ...
21/11/2019 Thursday 16:15 - 18:15 L26.12 Lecture Giacomo Boracchi CNN for localization and detection
22/11/2019 Friday 14:15 - 16:15 6.0.1 Lecture Giacomo Boracchi GAN
28/11/2019 Thursday 16:15 - 18:15 L26.12 Lecture Matteo Matteucci Recurrent Neural Networks
29/11/2019 Friday 14:15 - 17:15 6.0.1 Lecture Matteo Matteucci Vanishing gradient and LSTM
05/12/2019 Thursday 16:15 - 18:15 L26.12 Lecture Matteo Matteucci Word embedding
06/12/2019 Friday 14:15 - 17:15 6.0.1 Lecture Matteo Matteucci Variational Autoencoders
12/12/2019 Thursday 16:15 - 18:15 L26.12 -- -- -- No Lecture Today --
13/12/2019 Friday 14:15 - 17:15 6.0.1 Practicals Francesco Lattari Keras examples ...
19/12/2019 Thursday 16:15 - 18:15 L26.12 Lecture Matteo Matteucci ... TBD ...
20/12/2019 Friday 14:15 - 17:15 6.0.1 Lecture Matteo Matteucci ... TBD ...

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

Slides from the lectures by Giacomo Boracchi are avaible 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.

Additional material from the teachers

  • ... coming soon ...