📚 Curiosity Chronicles

Machine Learning Basics

2 min read

Machine Learning (ML) is a subset of artificial intelligence that focuses on developing algorithms and models enabling computers to learn from data. Unlike traditional programming, where explicit instructions are provided, ML systems learn patterns and make predictions or decisions without being explicitly programmed. ML plays a pivotal role in various applications, from image recognition to natural language processing, contributing to the advancement of AI.

4 Major Techniques of ML

ML encompasses different techniques tailored for specific tasks. The four major techniques are:

  • Supervised Learning - Uses labeled data for training.
  • Unsupervised Learning - Deals with unlabeled data and identifies patterns.
  • Semi-Supervised Learning - Combines both labeled and unlabeled data.
  • Reinforcement Learning - Involves learning through trial and error.
graph TD
    A(4 Major Techniques of ML) -->|Uses labeled data| B(Supervised Learning)
    A -->|Deals with Unlabeled data| C(Un-Supervised Learning)
    A -->|Combines labeled and unlabeled data| D(Semi-Supervised Learning)
    A -->|Learning through trail and error| E(Reinforcement Learning)

Understanding ML Techniques

Supervised Machine Learning

Involves labeled data with known inputs and correct outputs. It compares the predicted output to the correct outputs and tunes it to be more aligned with the correct output.

[[Data in ML#|Train, Validation and Test Datasets are involved]]

Algorithms based on supervisor machine learning

  • Regression algorithms
  • Logistic Regression algorithms
  • Classification algorithms
    • Naive Bayes classifiers
    • Decision Trees
    • Support vector machine

Unsupervised Machine Learning

Deals with unlabeled data and clusters it based on patterns. The output is not known for the given input. Used for Grouped Datasets, Anomaly Detection, association mining, Latent variable models

  • Low accurate than supervised
  • Spectral classes
  • Class interpretation (the classification should be understood by human)

Semi-Supervised Learning

Uses a combination of labeled and unlabeled data. It is cost-effective as obtaining labeled data can be challenging.

Reinforcement Learning

Learning through trial and error. Key components include Agent, Environment, and Actions.

Examples of ML Applications

ML finds applications in various domains, shaping the landscape of technology. Some notable examples include:

  • Anomaly Detection - Identifying unusual patterns or events.
  • Computer Vision - Teaching machines to interpret and make decisions based on visual data.
  • Natural Language Processing (NLP) - Understanding and processing human language.
  • Conversational AI - AI systems capable of engaging in natural conversations.

Explore ML Algorithms

Confusion Matrix

Graph View

Backlinks