Notes, demos and materials for learning Machine Learning

In addition to the material in this git repository, I've also used materials from my computer vision, data mining and deep learning modules. Please feel free to take a look at the lecture slides and notes for these which can be found here:

• http://comp3204.ecs.soton.ac.uk / https://github.com/jonhare/COMP3204
• http://comp6237.ecs.soton.ac.uk / https://github.com/jonhare/COMP6237
• http://comp6248.ecs.soton.ac.uk / https://github.com/jonhare/COMP6248
• http://ecs-vlc.github.io/COMP6258 / https://github.com/ecs-vlc/COMP6258

Weather permitting Jon, Niranjan and Adam are likely to go for a walk (possibly to a pub on occasion). If you would like to join us (or go for and independent walk) please take comfortable shoes.

(Note that this is only a guide. We'll adapt the content to your needs during the course.)

• Monday 1st July 2024: Overview of Machine Learning

  • Leaders: Prof Niranjan, Prof Prugel-Bennett and Prof Hare
  • 9:00-10:00 Available outside Voltaire/Berkeley meeting room
    • Tea & Coffee welcome
  • 10:00-10:30 (Voltaire/Berkeley)
    • Introductions: Course teachers and students
    • Chat in Break out rooms
  • 10:30-12:50 Niranjan
    • ML in one page Niranjan
    • Artificial Idiots Adam
    • Failures of machine learning Jon
  • 1:00-2:00
    • Lunch
  • 2:05-3:30 Niranjan
    • Understanding simple machine learning algorithms
      • Linear models, Gaussian distributions
    • Bayes Optimal Regression
    • Fisher Discriminant Analysis
    • Perceptron
    • Feature selection and Lasso
  • 3:30-4:00 Bar
    • Coffee
  • 7:15-9:00 Dinner

• Tuesday 2nd July 2024: Introduction to Machine Learning

  • 8:00-9:00 Breakfast
  • 9:00-10:30 Adam
    • How to do Machine Learning
  • 10:30-11:00
    • Coffee/break out rooms
  • 11:00-12:50 Jon
    • Handling Data
    • Hands-of practical session
    • Introduction to python, scikit-learn and CoLab
  • 1:00-2:00
    • Lunch
  • 2:00-3:30 Niranjan
    • MLPs
    • Gradient learning, SGD, momentum
    • valuating performance
      • ROC curves
  • 3:30-4:00 Bar
    • Tea & Coffee
  • 4:00-5:00
    • Ethics discussion
  • 7:00-9:00 Dinner

• Wednesday: 3rd July 2024: Advanced Machine Learning

  • Leader: Adam
  • 9:00-10:30
    • Generalisation
      • Bias-Variance Dilema
    • Kernel methods
      • SVM
      • kernels
  • 10:30-11:00
    • Coffee
  • 11:00-12:50
    • Ensemble Techniques
    • Bagging, random forest and Boosting
  • 1:00-2:00
    • Lunch
  • 13:00-3:30
    • Bayesian Inference
    • Probability Models
      • Gaussian Processes and Naive Bayes
  • 3:30-4:00
    • Coffee
    • Homework
    • Extra exercises: http://comp6248.ecs.soton.ac.uk/labs/lab1/
  • 7:00-9:00 Dinner_

• Thursday 4th July 2024: Deep Learning

  • *Leader: Jonathon
  • 9:00-10:30
    • Why Deep (see https://github.com/jonhare/DiffProgLecture/blob/main/tex/diffprog.pdf)
      • CNNs
      • RNNs (LSTM, etc.)
      • Current research challenges
      • Visual
        • segmentation
        • object detection
        • multi-label classification
      • Text
        • sequence-sequence learning
          • translation, embedding, etc
          • logical inference & QA
      • Cross-modal transfer
        • generating from embeddings
        • VQA
      • GANs
  • 10:30-11:00
    • Coffee
  • 11:00-12:50
    • Word Embeddings
    • Loss functions
    • GPU programming (libraries)
  • 1:00-2:00
    • Lunch
  • 1:30-3:00
    • Keras tutorial 1 - building simple CNNs
    • Transfer Learning
    • Keras tutorial 2 - transfer learning with CNNs
    • or: http://comp6248.ecs.soton.ac.uk
    • Keras tutorial 3 - Text classification
    • Keras tutorial 4 - Sequence modelling
    • or: http://comp6248.ecs.soton.ac.uk 3:00-3:30
  • 3:30-4:00
    • Coffee
  • 7:00-9:00 Dinner

• Friday 5th July 2024: Practical Machine Learning

  • Leaders: Prof Niranjan, Prof Prugel-Bennett and Dr Hare
  • 9:00-10:30
    • Workshop on data you provide
    • We will look at (slides):
      • Analyse the problem
      • Visualise the data
      • Cleaning the data
      • Using machine learning libraries
      • Evaluate performance
  • 10:30-11:00 Coffee
  • 11:00-12:50
    • Work on data
  • 12:30-1:30 Lunch
  • 13:00-3:30
    • Practical ML
  • 3:30-4:00 Coffee
  • Leave