Unsupervised learning explained

Posted on 07-08-2019 , by: admin , in , 0 Comments

Despite the success of supervised machine learning and deep learning, there’s a school of thought that says that unsupervised learning has even greater potential. The learning of a supervised learning system is limited by its training; i.e., a supervised learning system can learn only those tasks that it’s trained for. By contrast, an unsupervised system could theoretically achieve “artificial general intelligence,” meaning the ability to learn any task a human can learn. However, the technology isn’t there yet.

If the biggest problem with supervised learning is the expense of labeling the training data, the biggest problem with unsupervised learning (where the data is not labeled) is that it often doesn’t work very well. Nevertheless, unsupervised learning does have its uses: It can sometimes be good for reducing the dimensionality of a data set, exploring the pattern and structure of the data, finding groups of similar objects, and detecting outliers and other noise in the data. 

In general, it’s worth trying unsupervised learning methods as part of your exploratory data analysis to discover patterns and clusters, to reduce the dimensionality of your data, to discover latent features, and to remove outliers. Whether you then need to move on to supervised learning or to using pre-trained models to do predictions depends on your goals and your data.

What is unsupervised learning?

Think about how human children learn. As a parent or teacher you don’t need to show young children every breed of dog and cat there is to teach them to recognize dogs and cats. They can learn from a few examples, without a lot of explanation, and generalize on their own. Oh, they might mistakenly call a Chihuahua “Kitty” the first time they see one, but you can correct that relatively quickly.

Children intuitively lump groups of things they see into classes. One goal of unsupervised learning is essentially to allow computers to develop the same ability. As Alex Graves and Kelly Clancy of DeepMind put it in their blog post, “Unsupervised learning: the curious pupil,”

Unsupervised learning is a paradigm designed to create autonomous intelligence by rewarding agents (that is, computer programs) for learning about the data they observe without a particular task in mind. In other words, the agent learns for the sake of learning.

The potential of an agent that learns for the sake of learning is far greater than a system that reduces complex pictures to a binary decision (e.g. dog or cat). Uncovering patterns rather than carrying out a pre-defined task can yield surprising and useful results, as demonstrated when researchers at Lawrence Berkeley Lab ran a text processing algorithm (Word2vec) on several million material science abstracts to predict discoveries of new thermoelectric materials.