Neural Networks

1 Session 01

1.1 Definitions

  • Artificial Intelligence (AI): A program that can see, reason, act and adapt.
  • Machine Learning (ML): Algorithms whose performance improve as they are exposed to more data over time
  • Deep Learning (DL): Subset of machine learning in which multilayered neural networks learn from vast amounts of data

1.2 Machine learning

These programs learn from repeatedly seeing data, rather than being explicity programmed by humans.

1.2.1 Supervised Learning

  • Dataset: Has a target column
  • Goal: Make predictions
  • Example: Fraud Detection

1.2.2 Unsupervised Learning

  • Dataset: Does not have a target column
  • Goal: Find structure in the data
  • Example: Customer Segmentation

1.2.3 Limitations

  • Suppose you wanted to determine if an image if of a cat or a dog
  • What features would you use?
  • This is where deep learning can come in

1.3 Deep Learning

Is machine learning that involves using very complicated models called deep neural networks

1.4 Difference Classic ML and DL

1.4.1 ML

  • Define Features by ourselves, before we feed the data to our actual model
  • Only after defining features, we can feed them through model

1.4.2 DL

  • Will combine the 2-step process
  • From an input, will train the model to detect features

1.5 Quiz

1. Machine Learning is a subset of Artificial Intelligence

  1. False
  2. True

2. Deep Learning is a subset of Machine Learning

  1. False
  2. True

3. Marchine Learning consist in programming computers to learn from real-time human interactions

  1. False
  2. True