In this day and age there are booming buzz words in the tech sphere. Some being Artificial Intelligence, machine learning and deep learning.
What do these words mean? We'll cover them in a few but we'll put our focus majorly in machine learning.
Artificial Intelligence
According to a research paper by John McCarthy artificial intelligence is the science and engineering of making intelligent machines, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to methods that are biologically observable.
Turing on the other hand, provided a set of tests that an artificially intelligent computer has to pass.
Generally AI focuses on the simulation of human intelligence in computers/machines.
Machine Learning
It is a subset of artificial intelligence (AI) in which algorithms are developed and trained on a dataset from which they make predictions or decisions based on the in fed data.
This means that computers learn from data rather than being programmed.
There are three types of machine learning algorithms.
- Supervised Learning
- Unsupervised Learning
- Reinforcement Learning
a. Supervised Learning
The algorithms are trained with labeled datasets. This means that the training data has both the input features and the expected output.
It pretty similar to when you teach a class by also providing them answers for the test so that they can gauge their understanding.
Thus the algorithm learns to recognize patterns to make predictions/decisions about new, unlabeled data.
Some of the common supervised algorithms are:
- linear regression
- logistic regression
- random forest
- support vector machine(svm)
b. Unsupervised Learning
The algorithm is trained on unlabeled data. This means that the algorithm has to recognize patterns in the data and then group similar data together.
Some common algorithms used in unsupervised machine learning are:
- neural networks
- k-means clustering
- probabilistic clustering methods
c. Reinforcement Learning
The algorithm learns to make decisions by interacting with an environment and receiving feedback in the form of reward or penalties. The algorithms strives to maximize rewards while reducing penalties.
An example of a reinforcement learning algorithm is a robot to pick items. It holds items and attempts to lift them. If the item falls the robots via trial and error adjusts various features such as holding more tightly or looking for a more suitable place to hold the item. Over time it learns the best way to pick up the item and now significantly does so effortlessly.
Applications of Machine Learning
- self driving cars
- speech recognition
- computer vision
- recommendation engines
- fraud detection
Machine learning is an essential tool in data science. We'll explore these algorithms as we progress.
Top comments (0)