Applied Machine Learning - Week 4

Monday the 16th - Friday the 20th of May 2022

Monday 16th of May (afternoon):
Lectures: An introduction to Convolutional Neural Networks (JK).
     Very cool visualisation of how a CNN works on your written numbers.
     Recording of Last years CNN lecture video (112 MB).

Exercise: First step is a small introduction to PyTorch, simply to make everyone familier with the basic code parts (don't spend too long on this)
     The main exercise consists of classifying tephra (i.e. volcanic ash) from Greenlandic ice cores, based on images of the tephra.
     For inspiration, see how to recognise handwritten numbers with a Convolutional Neural Network: CNN_MNISTdata.ipynb.
     This data can (also) be obtained from (famous) Yann LeCun's webpage: Link to MNIST dataset.


Wednesday 18th of May (morning):
Lectures: Recurrent Neural Networks (RNN), Long Short Term Memory (LSTM) and Natural Language Processing (NLP) (JK).
     Recording of Last years RNN/LSTM lecture video (222 MB).

Exercise: First, consider the example LSTM making FlightPassengerPredictions.ipynb on 1949-1961 airline data.
     Then, use an LSTM to do Natural Language Processing, in this case estimate if a review was good or bad based on IMDB data.


Wednesday 18th of May (afternoon):
Lectures: Auto-Encoder and anomaly detection (TP).
     AutoEncoders at work: Detecting anomalies with AutoEncoders (Carl Johnsen).
     The ImageNet Competition - a revolution in ML (TP).

Exercise: The exercise is about classifying digits i.e. the MNIST data, but this time with an autoencoder in an unsupervised manner.

---

Last updated: 15th of May 2022 by Troels Petersen.