Machine Learning & Deep Learning
Deep Learning has in recent years revolutionized research in machine learning and led to AI receiving renewed attention.
In this lecture you will learn how to get started and use artificial neural networks and other deep learning techniques.
The course requires knowledge in programming, it is also an advantage with prior knowledge in data science
Innehåll: Machine Learning & Deep Learning
Birger Moëll Machine Learning Research Engineer KTH / Ayond
09:00-09:30 Introduction to Machine learning / Deep Learning | Talk |
Introduction to Machine Learning Slides
09:30-09:45 Getting your machines ready for machine learning | Code |
Write code for installing properly on mac and windows
Install for windows and mac, keras, tensorflow, numpy.
09:45-10:00 Coffee and break
10:00-10:15 Hello World in Machine Learning (MNIST) | Talk |
Code for MNIST. Explanation of MNIST
10:15-10:45 Running your own MNIST | Code |
Getting your computer working with MINST
Exploring MNIST and running different models for solving MNIST
10:45-11:00 Coffee and break
11:00-11.15 Feedforward Neural Networks | Talk |
What is a neural network, activation functions, math behind, neuroscience
11.15-11.45 Building your own feedforward neural network | Code |
Build a neural network to handle data from neuroscience (the data is processed, you just need to build the network)
12:45-13:00 Q and A | Interactive
13:00-13.15 Image recognition and convolutional neural networks | Talk |
Slides regarding image recognition, how it works, neuroscience, math
13:15-13:45 Building your own convolutional neural network | Code |
Training a classifier of cat vs dog
13:45-14:00 Coffee and break
14:00-14.15 Time series prediction and LSTMs | Talk |
Slides regarding time series data and LSTMs, What are LSTMs useful for.
14:15-14:45 Building your own LSTMs | Code |
Borrow from the unreasonable effectiveness of LSTMS, jupyter notebook for working with LSTM data.
14:45-15:00 Coffee and break
15:00-15.15 Generative models | Talk |
Talking about generative models, how can they be used.
15:15-15:45 Trying out GANS | Code |
Code for style transfer? Generative models
15:45-16:00 Coffee and break
16:00-16.15 Machine learning in the wild | Talk |
How to host your models, Flask, Google, AWS, Azure, Tensorflow.js
16:15-16:45 Serving your own machine learning model | Code |
Building your own flask model to train.
16:45-17:00 Q and A | Interactive
Kursslut (ca): 17.00