Basic Terminology in Artificial Intelligence
Welcome to our blog posts, we are here to make you understand the basics of Artificial intelligence, for this matter, we will be summarizing the book called "Applied Artificial Intelligence". This book starts by giving a brief definition for many terms that will be useful throughout the book. We added all the definitions to this blog for you to be able to understand the terms along this blog.
Artificial General Intelligence
Refers to machines with human-level or higher intelligence, who are capable of abstracting concepts from limited experience and converting them into knowledge between domains.
AGI is also known as Strong AI which refers to a system designed for one specific task and whose capabilities are not easily copied or transferable to other systems.
Statics
Defined as the discipline concerned with the collection, analysis, description, visualization, and drawing of interferences from data. Its main focus is on describing the properties of a dataset and the relationships that exist between data points.
Descriptive statics
Describes the basic features of the data that is being studied.
Inferential statistics is used to draw conclusions that apply to more than just the data being studied.
Data mining is the automation of exploratory statistical analysis on large-scale databases. The goal is to extract patterns and knowledge from large-scale datasets for them to be reshaped into a more understandable structure for later analysis.
Symbolic systems are programs that use human-understandable symbols to represent problems and reasoning. The best and most successful form of symbolic systems is the expert system which mimics the decision-making process of human experts. This is more effective when applied to automated calculations and logical processes where rules and outcomes are relatively clear.
Machine learning enables computers to learn without being explicitly programmed.
Supervised learning is when the computer is given labeled training data, which consists of paired inputs and outputs, and learns general rules that can map new inputs to the correct output. Commonly used for classification.
Unsupervised learning occurs when computers are given unstructured rather than labeled data and asked to discover inherent structures and patterns that lie within the data.
Semi-supervising learning lies between supervised and unsupervised learning. Active learning is used to optimize recommendation systems.
Reinforcement learning is learning by trial and error, where a computer program is instructed to achieve a stated goal in a dynamic environment.
Deep learning is a subfield of machine learning that builds algorithms by using multi-layer artificial neural networks.
Probabilistic programming enables us to create learning systems that make decisions in the face of uncertainty by making inferences from prior knowledge.
Ensemble methods
Combine different machine learning models or blend deep learning models with rule-based models
Categories
Bagging: entails training the same algorithm on different subsets of the data.
Boosting: Involves training a sequence of models where each model prioritizes learning from the examples that failed before.
Stacking: pull the output of many models.
Bucketing: Train multiple models for a given problem and dynamically choose the best one for each specific input.
WBE (Whole Brain Uploading): Known as mind uploading, seeks to replicate the human level of intelligence in machines by digitalizing human brains.
Source: Yao, M., Jia, M. and Zhou, A., 2018.Applied Artificial Intelligence. pp.219-381.
