June 1, 2020 Deepak Raj
In 1959, Arthur Samuel, a pioneer in the field of machine learning (ML) defined it as the “field of study that gives computers the ability to learn without being explicitly programmed”.
The algorithm is feed with example input and their desired outputs. The goal is to learn a general rule that maps input to outputs. The training process continues until the model achieves the desired level of accuracy on the training dataset.
The goal is the predicted output for new input. E.g.
Unsupervised machine learning is opposite to the supervised. It doesn't have label data and more accurate to the real data world database. In the real-world databases are rarely labeled. No label is given to the learning algorithms, leaving it's own to find a relationship and give output.
The algorithms work trial and error to come up with a solution to the problem. The algorithms get rewards or penalties on their action to improve. It's like teaching a child and giving him reward on the right answer and punishment on the wrong answer. it's the goal to minimize penalties. it's up to the model to find out how to perform the task. Reinforcement learning is the key to true Artificial intelligence.
Eg. Training the model to control the car is an example of reinforcement learning. The model has to minimize the risk factor and improve safety, minimize time ride, reduce pollution, obey traffic rules, passenger's safety and comfort.
Machine Learning is a trendy topic nowadays. The field of ML is experiencing exponential growth. ML algorithms are getting more accurate day by day. Algorithms are already better than humans in recognizing and analyzing images.
The progress we’ve made from 26% error in 2011 to 3% error in 2016 is hugely impactful. The way I like to think is, computers have now evolved eyes that work.
If you like to write article or Improve this article for CodePerfectPlus mail us at firstname.lastname@example.org
tags: machine learning, data science