Machine Learning Practical (MLP) Lectures 2020-21

Machine Learning Practical (MLP) Lectures 2020-21

The Machine Learning Practical (MLP) course for 2020-21 is concerned with deep neural networks. Doing this course involves the following:

  • Implementing deep learning systems using python;
  • Training and evaluating on data sets for tasks such as handwriting recognition;
  • Designing and running machine learning experiments to investigate research questions;
  • Reporting on your experiments, discussing and interpreting the results.

During semester 1 we shall investigate neural network learning with a focus on the classification of handwritten digits using the well-known MNIST dataset. Using a Python software framework that we shall provide, and a series of Jupyter notebooks, the course will explore multi-layer neural network classifiers, convolutional network classifiers, and recurrent networks. The lectures in semester 1 will provide the required theoretical support for the practical work.

Semester 2 will be based around group projects, typically using TensorFlowPyTorch, or another deep learning toolkit. The lectures in semester 2 will cover more advanced material in deep learning.

Playlists:

Single-Layer          Deep NNs             Deep NNs              CNNs                     RNNs

Lecture 01             Lecture 03             Lecture 05             Lecture 07             Lecture 09

Lecture 02             Lecture 04             Lecture 06             Lecture 08             Lecture 10

Q&A 1                    Q&A 2                   Q&A 3                    Q&A 4                   Q&A 5     

Advanced NN Tutorials [Relational Networks, GANs, HyperNetworks, Attention, Transformers, Meta-Learning]

Google Cloud tutorial

MLP Interviews - Winter 2020 [Jinhong Lu, Leonie Bossmeyer, Maximiliana Behnke, Maysara Hammouda]


Guest Lectures:

Week 2 - Antreas Antoniou (UoE) - MLP Cluster

Week 3 - Ben Allison (Amazon) - Building Production Machine Learning Systems

Week 4 - Sofia Vallecorsa (CERN) - Generating models for High Energy Physics

Week 5 - Oisin Mac Aodha (UoE) - ML for Computer Vision

Week 6 - Mirella Lapata (UoE) - ML for Machine translation

Week 7 - Siddharth Narayanaswamy (UoE) - Explainability in ML

Week 8 - Sanjay Rakshit (Transwap) - Framework for Implementing an ML Project: from Concept to Production

Week 9 - Daniel Angelov and Svetlin Penkov (Efemarai) - Visualising and Testing Deep Learning Computations

Week 10 - Puneet Dokania (Five AI) - Continual Learning

…Read more Less…

 Public, Restricted

77 Media
4 Members
Managers:
Appears In: