DEV Community

An Vo
An Vo

Posted on

AI Learning Journey #1: What Is AI?

#genai #ai #machinelearning

AI often feels mysterious, but it starts with a simple idea: machines learning from data.

What is AI?

  • Humans learn from experiences, and AI learns by processing vast amounts of data to make informed decisions.
  • Data is indispensable for AI as it aids in learning, pattern recognition, and making informed decisions or predictions.

Foundational AI Concepts and Terminology

  • Artificial Intelligence (AI): The field of building machines that can make decisions or perform tasks in ways that resemble human intelligence.
  • Machine Learning (ML): a subset of AI, uses data to train the computer. The computer will learn and improve over time.

  • Supervised Learning: A learning process where models are trained on labeled data (data with correct answers). For example, a model can learn to classify cats and dogs after being trained on labeled images.
    Supervised Learning

  • Unsupervised Learning: is the process that uses big data without labels to train the model, it will learn and find patterns by itself. For example, Unsupervised methods can analyze physiological and behavioral data to identify mental health conditions, especially when labeled data is in short supply.

  • Reinforcement Learning: if the model returns a correct answer, then send a reward to it. If the answer is incorrect, it receives a penalty.

  • Deep Learning A subset of machine learning based on neural networks. Common architectures include:

    • Artificial Neural Networks - Inspired by the human brain, they can process massive datasets and learn patterns from them.
    • Convolutional Neural Networks (CNNs): used in vision tasks, can extract context from images.
    • Recurrent Neural Networks (RNNs): Efficient for sequential data and are often used for predicting text. RNNs are adept at handling time-series predictions in stock prices.

  • Generative Models: Models that learn from existing data and generate new, similar data samples, making them useful for creating content like text, images, or music.

  • Transformer Architecture: Consisting of encoders and decoders, crucial in language translation and code generation tasks.

Transformer Architecture

This post lays the foundation for my AI learning journey. In the next post, I’ll dive deeper into AI Learning Journey #2: What Does “Training a Model” Mean?

I’m learning AI step by step with the goal of preparing for AWS Certified Generative AI Developer – Professional (AIP-C01) certification, so this series will build toward that goal.
https://docs.aws.amazon.com/aws-certification/latest/examguides/ai-professional-01.html
Checklist:
❑ AI basics
❑ ML fundamentals
❑ Foundation for GenAI concepts

Top comments (0)