Machine learning concepts

Understanding Machine Learning

Lis Sulmont

Curriculum Manager, DataCamp

Three types of machine learning

$$

1) Reinforcement learning

2) Supervised learning

3) Unsupervised learning

Training data

• Training data: existing data to learn from
• Training a model: when a model is being built from training data
  • Can take nanoseconds to weeks

Supervised learning training data

Supervised learning training data

Supervised learning training data

Supervised learning training data

Supervised learning training data

After training (supervised learning)

After training (supervised learning)

After training (supervised learning)

Supervised vs unsupervised learning

• Supervised learning
  • Training data is "labeled"
• Unsupervised learning
  • Training data only has features
  • Useful for:
    • Anomaly detection
    • Clustering, e.g., dividing data into groups

Unsupervised learning training data

Unsupervised learning training data

After training (unsupervised learning)

Unsupervised Learning

• In reality, data doesn't always come with labels
  • Requires manual labor to label
  • Labels are unknown
• No labels: model is unsupervised and finds its own patterns

Let's practice!

Understanding Machine Learning