Concluding remarks

Designing Machine Learning Workflows in Python

Dr. Chris Anagnostopoulos

Honorary Associate Professor

Concluding remarks

  • Refresher of supervised learning pipelines:
    • feature engineering
    • model fitting
    • model selection
  • Risks of overfitting
  • Data fusion
  • Noisy labels and heuristics
  • Loss functions
    • costs of false positives vs costs of false negatives
Designing Machine Learning Workflows in Python

Concluding remarks

  • Unsupervised learning:
    • anomaly detection
    • novelty detection
    • distance metrics
    • unstructured data
  • Real-world use cases:
    • cybersecurity
    • healthcare
    • retail banking
Designing Machine Learning Workflows in Python

Congratulations!

Designing Machine Learning Workflows in Python

Preparing Video For Download...