Deep learning vs machine learning

This year 2020 we are celebrating 70 years of the Artificial Intelligence concept . It all started when the British mathematician Alan Turing has published his paper in a Philosophical journal called Mind. The paper questioned us about the possibility of a thinkable machine. A machine that could imitate the human behaviors. The idea behind of this type of machine was so revolutionary that we are still struggling today to replicate the "perfect model" and it will continue for years to come. Can you imagine how deep is the following question: If a computer could imitate the sentient behavior of a human would that not imply that the computer itself was sentient?

Machine Learning

Machine learning (ML) and Deep learning (DL) are both subsets of Artificial Intelligence.

ML are “Algorithms that parse data, learn from that data, and then apply what they’ve learned to make informed decisions” An easy example of a ML algorithm is an on-demand music streaming service. For the service to make a decision about which new songs or artists to recommend to a listener, machine learning algorithms associate the listener’s preferences with other listeners who have a similar musical taste.

It involves the creation of algorithms which can modify itself without human intervention to produce desired output- by feeding itself through structured data. ML algorithms are built to “learn” to do things by understanding labeled data.

Deep Learning

DL is a recent field that occupies the much broader field of Machine Learning. Deep Learning is most famous for its neural networks such as Recurrent Neural Networks, Convolutional Neural Networks, and Deep Belief Networks. While other machine learning algorithms employ statistical analysis techniques for pattern recognition, Deep learning is modeled after the neurons of the human brain.

Being short, deep learning is considered an evolution of machine learning. It uses a programmable neural network that enables machines to make accurate decisions without help from humans.

In order to understand deep learning, we have to understand how the nervous system in the human body works. As we all know that our nervous system is built up of neurons. These neurons are able to grasp information that is transmitted to our body. These neurons have the ability to learn information over time and consequently make decisions.

Deep Learning Applications 

There are several fields for DL, such as Cancer Diagnostics by using photos of cell tissues, self-driving car industry by identifying obstacles, car plates, lane problems, etc., E-commerce by self-feeding customers based on their purchase, Industrial Equipment inspection by trained models that can identify quality issues or defective equipment or products, Predictive maintenance inspection by machine data collected from end sensor nodes to draw meaning insight to predict machine failures. The list is very extensive where the limitations lie usually on the investment on computer to process the data and as well the time required to process the initial learning.

In machine learning, feature engineering is a fundamental job as it improves accuracy and sometimes the process can require domain knowledge about a certain problem. One of the biggest advantages of using deep learning approach is its ability to execute feature engineering by itself. In this approach, an algorithm scans the data to identify features which correlate and then combine them to promote faster learning without being told to do so explicitly.

There are several advantages of using deep learning algorithms specially with IoT and 4.0 Industry.

These days, deep learning has come a long way from being just a trend and it’s quickly becoming a critical technology being adopted steadily by an array of businesses, across multiple industries. Unlike 1950 time, today more companies are investing and developing technologies and infrastructure that can support a more robust Deep learning algorithm. It makes the AI implementation process less costly and with a faster response.

The main challenge for us mathematicians and data scientists is continuing improving and optimizing the Deep Learning algorithms so it can bring transparency on DL decisions for us human beings.

Juliano Santos Lima