AI Taxonomy

AI (Artificial Intelligence) can be broadly classified into three categories based on its functionality:

  • Reactive Machines: These are the most basic types of AI systems that do not have any memory or past experience. They can only react to the current situation based on pre-programmed rules. Examples of reactive machines include chess-playing computers, voice recognition systems, and self-driving cars.
  • Limited Memory: These AI systems have the ability to store some data and past experiences to make better decisions. They can use this stored information to make predictions and learn from their past experiences. Examples of limited memory AI systems include speech recognition software, personal assistants like Siri and Alexa, and recommendation engines.
  • Self-Aware AI: These are advanced AI systems that can not only learn from past experiences but also have the ability to reason, understand, and make decisions like a human being. These AI systems are capable of recognizing their own limitations and can adapt to new situations on their own. They can also analyze and process large amounts of data to make informed decisions. Self-aware AI systems are still in the research and development phase and are not yet widely available.

Apart from the above-mentioned categories, AI can also be classified based on its application or domain, such as natural language processing, computer vision, robotics, and so on.

Artificial General Intelligence (AGI)

AGI stands for Artificial General Intelligence, which refers to the development of intelligent machines that have human-level cognitive abilities across a wide range of tasks and domains. Unlike current AI systems that are designed for specific applications and tasks, AGI aims to create machines that can perform any intellectual task that a human being can do.

The development of AGI is considered to be a significant milestone in the field of AI as it would bring machines closer to human-level intelligence and pave the way for more advanced and sophisticated AI applications. AGI systems would be capable of learning and adapting to new situations, making complex decisions, and solving problems independently, without requiring explicit programming or human intervention.

However, achieving AGI is still a significant challenge for the AI community, as it requires the development of new algorithms and techniques that can replicate the broad range of cognitive abilities that humans possess. Researchers are actively working on developing AGI, and several organizations and companies have dedicated resources to this area of research, including OpenAI, DeepMind, and IBM Watson, among others.

AI Taxonomy.txt
Machine Learning (ML)
  Deep learning
Natural Language Processing (NLP)
  Content extraction
  Machine translation
  Text generation
Expert Systems
  Image recognition
  Machine vision
  Speech to text
  Text to speech
A tour of machine learning algorithms approach taxonomy.txt
Supervised Learning
      Markov Process (MDP)
      Hidden Markov Model (HMM)
      Markov Random Fields
      Naïve Bayes (NB)
      Latent Dirichlet Allocation
      Belief Network
      Linear Regression
      Logistic Regression
      Neural Networks (ANN)
      Support Vector Machine (SVM)
      Maximum Entropy
      Decision Trees
      Conditional Random Fields (CRF)
      Random Forests
Unsupervised Learning
    K-means Clustering
    Spectral Clustering
    Hierarchical Clustering
    Expectation-Maximization (EM)
  Dimension Reduction
    Principal Component Analysis (PCA)
    Linear Discriminant Analysis (LDA)
Reinforcement Learning
    Model Based
      Iterative Value
      Iteraive Policy
    Temporal Difference
      Learning Classifiers
      Stochastic Gradient
      Genetic Algorithm
  • ai/taxonomy.txt
  • Last modified: 2023/03/25 20:32
  • by Henrik Yllemo