One of the key differences between humans and computers is that humans learn from past experiences, at least try, but computers or machines need human instructions to be told what to do.
That means if we want them to do something, we have to provide them with detailed, step-by-step instructions on exactly what to do. It makes machine to improve at tasks with experience. These set of instructions are nothing but programs.
No doubt! Machine learning has proven to be one of the most technological advancements of the past decade. In this increasingly competitive world ML/AI is helping companies to fast-track digital transformation & move into an age of automation. Before we get to know more about Machine learning, lets understand what is Artificial intelligence!
Artificial Intelligence
Artificial intelligence is defined as a program that exhibits cognitive ability similar to that of a human being. Making computers think like humans and solve problems the way we do is one of the main tenets of artificial intelligence. AI exists as an umbrella term that is used to denote all computer programs that can think as humans do. Any computer program that shows characteristics, such as self-improvement, learning through inference, or even basic human tasks, such as image recognition and language processing, is considered to be a form of AI.
Machine Learning
Machine learning is subset of artificial intelligence in which machines learn & improve from experience without being explicitly programmed.
There are different ways to train machine learning algorithm, each with their own pros & cons. To understand the pros & cons we must understand the data that these algorithms are trained on. There are 2 kinds of data.
1)Labeled Data 2)Unlabeled Data
Labeled data is nothing but data which has both Features & Target variable defined in it. It requires human efforts to label the data, to begin with. Unlabeled data only has one or none of the parameters in machine readable form. This negates the need for human efforts to label the data but requires more complex solutions.
Types of Machine Learning
- Supervised Learning
- Unsupervised Learning
- Reinforcement Learning
1) Supervised Learning
Supervised machine learning is one of the basic machine learning techniques in which the algorithm is trained on labelled data. Even though the data needs to be labelled accurately, this technique can be very power-full when used in the right circumstances. In supervised learning, the ML algorithm is provided with small labeled data, sample data. This data is the smaller representation of the bigger data, population data. This sample gives the algorithm an idea about the type of data, patterns & data points to be dealt with.
The algorithm then finds the relationship between the parameters given, essentially finding a cause & effect relationship between the parameters. At the end of the training, the algorithm has an idea of how the data works and the relationship between the input and the output. This algorithm is then deployed with the final dataset. This means the supervised machine learning algorithm continues to improve even after deployment, discovering new pattern & relationship by ingesting new data. Supervised Learning can be divided into 2 major groups
1)Regression 2)Classification
2) Unsupervised Learning
Unsupervised learning has an advantage over the supervised learning as it can find the hidden pattern in unlabeled data. This means that human labor is not required to make the dataset machine-readable, allowing much larger datasets to be worked on by the program. However, unsupervised learning does not have labels to work with, resulting in the creation of hidden structures. Relationships between data points are perceived by the algorithm in an abstract manner, with no input required from human beings. The creation of these hidden structures is what makes unsupervised learning algorithms versatile. Instead of a defined and set problem statement, unsupervised learning algorithms can adapt to the data by dynamically changing hidden structures. This offers more post-deployment development than supervised learning algorithms.
Best example of Unsupervised learning is Clustering
3) Reinforcement Learning
Reinforcement learning takes inspiration from how the humans learn in life. Here the algorithm tries to learn by its own & learn from new situations using trial & error method. Algorithm will be rewarded if the outcome is favorable else it will be penalized.
Reinforcement learning works by putting the algorithm in a work environment with an interpreter and a reward system. In every iteration of the algorithm, the output result is given to the interpreter, which decides whether the outcome is favorable or not.
In case of the program finding the correct solution, the interpreter reinforces the solution by providing a reward to the algorithm. If the outcome is not favorable, the algorithm is forced to reiterate until it finds a better result. In most cases, the reward system is directly tied to the effectiveness of the result.