Machine learning is helping computers to tackle upon the tasks that have previously been only done by people. Either it’s driving cars or translating text from one language to another, either it’s making predictions about weather etc. have been enable by Machine Learning. Nowadays predictions made about Weather, Rain etc. have achieved great accuracy thanks only to Machine Learning.
But for a person unaware of Machine Learning the question is what’s exactly machine learning and what is making the current boom in machine learning possible?
What is Machine Learning?
Broadly speaking, Machine Learning is the process of teaching a computing system to make predictions based on data which it is fed specifically called training data.
The term prediction can be as broad from predicting whether there are people crossing road by self driving car, whether fruits are in condition or not just looking at their pictures, recommending videos on platforms like Youtube based on previously watched videos and many more cases in which machine learning can be deployed.
The key difference between traditional computer systems and new ways of using Machine Learning is that the code is not hard coded to make prediction, rather code is used to make a model of a situation(roads, people, traffic lights etc. in case of self driving cars) which is given a lot of data and based upon that makes predictions.
Whats difference between Machine Learning and AI?
Alongside machine learning, there are various other approaches used to build AI systems, including evolutionary computation, where algorithms undergo random mutations and combinations between generations in an attempt to “evolve” optimal solutions, and expert systems, where computers are programmed with rules that allow them to mimic the behavior of a human expert in a specific domain, for example an autopilot system flying a plane.
Machine Learning is one the prominent ways using which companies are trying to achieve Artificial Intelligence. At the point of birth of AI as a field in 1950s, it was defined as capability of performing a task which otherwise requires human intelligence.
AI systems generally have following characteristics: –
- Learning from data
- Reasoning about a situation
- Problem solving
- Manipulation of data
- Social Intelligence
Along with Machine learning other approaches like evolutionary computation can also be used for mimicking the human behaviour, it’s done in Airplane’s auto-pilot systems.
What are the main types of Machine Learning?
Broadly following are the types of machine learning: –
- Supervised Machine Learning
- Unsupervised Machine Learning
- Semi-Supervised Machine Learning
- Reinforcement Machine Learning
Supervised Machine Learning
This approach just involved teaching a machine by giving it a lot of examples. That’s why this method is called learn by example. Training these systems is not an easy task it requires ton load of data, some systems have to be exposed to millions of examples for making better predictions.
As a result, the datasets used to train these systems can be vast, with Google’s Open Images Dataset having about nine million images, its labeled video repository YouTube-8M linking to seven million labeled videos and ImageNet, one of the early databases of this kind, having more than 14 million categorized images. The size of training datasets continues to grow, with Facebook recently announcing it had compiled 3.5 billion images publicly available on Instagram, using hashtags attached to each image as labels. Using one billion of these photos to train an image-recognition system yielded record levels of accuracy — of 85.4 percent — on ImageNet’s benchmark. The laborious process of labeling the datasets used in training is often carried out using crowdworking services, such as Amazon Mechanical Turk, which provides access to a large pool of low-cost labor spread across the globe. For instance, ImageNet was put together over two years by nearly 50,000 people, mainly recruited through Amazon Mechanical Turk. However, Facebook’s approach of using publicly available data to train systems could provide an alternative way of training systems using billion-strong datasets without the overhead of manual labeling.
Unsupervised Machine Learning
Unsupervised machine learning systems uses certain algorithms which can perform classification of data by finding patterns in it, which otherwise are not quite visible to human eye.
For example – Google News uses Unsupervised Machine Learning for clustering the news as per categories.
Semi-Supervised Machine Learning
As the name suggests, the approach mixes supervised and unsupervised learning. The technique relies upon using a small amount of labelled data and a large amount of unlabelled data to train systems. The labelled data is used to partially train a machine-learning model, and then that partially trained model is used to label the unlabelled data, a process called pseudo-labelling. The model is then trained on the resulting mix of the labelled and pseudo-labelled data. The viability of semi-supervised learning has been boosted recently by Generative Adversarial Networks ( GANs), machine-learning systems that can use labelled data to generate completely new data, for example creating new images of Pokemon from existing images, which in turn can be used to help train a machine-learning model.
Reinforcement Machine Learning
Basically it’s learning by hit and trail method. The most common example of this is Playing of a game like Chess by a computer.