What is Ai
Artificial Intelligence (Ai) is the term used for the technology of mimicking human intelligence with machines attempting to replicate our decision making, problem solving, learning nature, and even emotions. Ai is the behind the scenes of a machine and it can only work in its data set given by the device. As it learns and completes tasks, Ai can expand its data set and optimize its process to become more efficient. Although Ai has vast potential benefits that can help society, there are certain ethical and theological concerns that are critical to its development.
Machine Learning
Machine learning is a subfield of Ai that focuses on the machines ability to imitate human behavior, specifically our problem solving abilities. First described by Arthur Samuel as “the field of study that gives computers the ability to learn without explicitly being programmed.” This definition demonstrates the use of machine learning because of the difficulty to tell a program how to complete a complex task. While humans only need basic instructions for a task, a machine needs a detailed step-by-step process to follow that would be very time consuming. Machine learning helps with this because the program is given a data set that it can use and the programmer gives a model with parameters and weighting of different results so the machine can optimize itself. There are three types of machine learning, supervised, unsupervised, and reinforcement. Supervised learning has labeled data that the programmer shows the Ai to train it to recognize and associate different labels with certain data. Unsupervised learning is where the machine is given an unlabeled data set and it’s job is to find general patterns. Reinforcement learning is a trial-and-error process with a reward system for when the machine does something right. This is defined by the programmer to get a system that does a specific task.
Neural Networks
A neural network is a learning program that makes decisions similar to how the human brain does. It has artificial neurons (nodes) with their own weight and threshold in order to process the information and send it to the next layer. In a regular network, there is an input layer, some hidden layers, and an output layer. If the threshold is not met on one of these layers, then the data is not passed on to the next layer.
Deep Learning
Deep learning is the latest branch of Ai that uses machine learning techniques while using the architecture of neural networks. It requires much more data, processing power, and time to train than regular machine learning because it is made for complex problem solving. The best part of deep learning is how accurate and versatile it is. The only limiting factor is the amount of data it has access to, but with enough it can be scaled for massive projects and be able to do a variety of tasks. The program goes through the training process of machine learning, adjusting the threshold required for each node to pass on the data. It differs from regular neural networks because it has more hidden layers between the input and output layers and implements the machine learning process.
