Artificial Intelligence and Machine Learning

1. Develop in the field of machine learning.

2. Learning ML terminologies, when to use ML, and some of its benefits.

Introduction.

1. AI is the broad term for teaching machines to think

   • AI: AI programs try to understand and replicate human intelligence and learn from outcomes.

   • ML: a subset of AI where the computer program derives rules from the data itself. It tries to uncover patterns that are hidden in the data.

   • DL: a subset of ML where the computer program dynamically learns by adjusting actions based on continuous feedback to maximize a reward.

2. Terminology:

   • Model: trained algorithm that you use to identify patterns in the data.

   • Training Algorithm: use data to train algorithms to make generalizable inferences.

   • Computer Vision.

   • Artificial Neural Networks.

   • Compute Model Instances: AWS provides a variety of instance types that support ML models, depending on the use case of the model.

   • Inference: the process of making predictions using a trained model.

Classification and Regression Problem Types

1. Binary Classifications:

   • The goal of training a binary ML model is to predict values that can only have one of two states, such as True or False.

   • Ex: Classify an observation into one of two categories to determine whether the variable is fraudulent or not.

2. Multiclass Classifications:

   • Classify an observation into one of three or more categories.

   • Ex: A model predicts why a customer is calling your store.

3. Regression:

   • A model maps inputs to a continuous value, like an integer.

   • Ex: Predict the stock price in a specific date in future.

Classical Programming vs. Machine Learning

1. Classical Programming:

   • Create rules based on factors like business requirements and domain knowledge.

   • Ex: Purchase product X ==> show customer product Y.

2. Machine Learning:

   • Derives rules from the data itself. Use a variety of data collected in the past to uncover patterns hidden in that data.

   • Ex: Recommendation for each user.

Algorithm Types:

1. Supervised Learning.

2. Unsupervised Learning.

3. Reinforcement Learning.