Machine learning (ML) and artificial intelligence (AI) have a present and future impact on every aspect of our business and personal lives. Also, there is a level of confusion about the ML, as there is no clear and shared understanding of what reasonable expectations for ML/AI are and how its success can be measured.
Everyone needs a basic understanding of ML/AI. Software developers must cut through the marketing diversions of software, hardware and services vendors in order to match up to the organization's ML strategy.
We all need to understand what ML and AI are. While ML/AI requires complex algorithms to work, it is not difficult to understand and apply those algorithms. We can take advantage of what exists in ML/AI today and plan for more exciting things in the near future.
ML/AI leverages mathematical algorithms to translate a set of input variables into concrete predictions. The ML model can make valid decisions based on input variables it has never seen before.
The self-driving car is a basic example for it. The self-driving car will avoid hitting pedestrians in any situation and any parameters. To train a machine to drive a car took a multi billion dollar investment that did not result in an AI model that applies to other problems.
In other words, to train the ML model to be able to drive a car was a long process of solving separate challenges, from the pedestrian safety rules to how it decides on speed when a sign isn't visible. Many of these situations have more than one dimension and significant legal implications if addressed incorrectly.
To train this self-driving car , we take the machine for a ride. We take it for hundreds of thousands of rides with all different drivers in different cars on different roads during different times in different countries. We fill up the car with sensors such as Lidar, cameras, distance sensors, and measures all inputs from the driver so that the machine can observe as many natural traffic situations as possible.
This unsupervised learning approach is not an easy task, and it requires a lot of expensive experimental vehicles and process power. But, it alone is not sufficient. It does not give the machine enough relevant feedback on which driving actions it should learn to imitate, versus the ones it should avoid. In an unsupervised learning approach, the machine will mostly attribute negative ratings to behavior that led to an actual incident, such as a crash.
Thus, it would be unethical to allow the machine to run over pedestrians for learning purposes, but by simple human driver observation, the ML/AI software would never receive enough experience in this specific department. Other methods, such as using a simulator or showing images of accidents to the machine, are only supplementary measures that do not replace the analysis of actual live sensor data and event feedback.