Artificial intelligence - a collection of multiple technologies that enable machines to sense, analyze, respond, and even learn on their own to augment human activities. AI is a new factor of growth in society that has the potential to drastically improve production in business and efficiency for the everyday lives of humans.
THE BASICS - Unbeknown to most, computer scientists such as Alan Turing and Marvin Minsky began developing AI 70 years ago. In general, the term artificial intelligence refers to multiple technologies that can be combined in different ways to sense, comprehend, and respond. Consider digital visual and audio processing. This technology can actively “sense” the environment by analyzing images, sounds, and speech. So, how exactly are these AI capabilities designed to learn from experience and adapt to the human world?
MACHINE LEARNING - There’s some lingering confusion on the functionality of machine learning and deep learning.
Both components are utilized in creating AI technologies but in different ways. By definition, machine learning1 involves data and output running on the computer to automatically create a program. This differs from traditional programming where data and a program are ran on the computer to produce an output. Machine learning teaches computers to program themselves, allowing the data to do the work rather than people. There are several types of machine learning like supervised, unsupervised and reinforcement learning. Huge amounts of data are collected everyday and machine learning helps programming all that data measurable.
DEEP LEARNING - Deep learning has produced many useful cases of machine learning and AI. Cases like driverless cars and preventative healthcare have been made possible because of the way deep learning breaks down tasks that assist the machine. Deep learning refers to the creation of artificial neural networks where each network has discrete layers, connections, and directions of data propagation. Each neuron in a network is assigned a weighting to its input and the final output is determined by the total of those weightings. The accuracy of the input is relative to the task being performed and the amount of data points that can be analyzed. Thanks to deep learning, the expansion of artificial intelligence continues.
IDENTIFY THE PROBLEM FIRST - One expert suggests that it may be useful to look at AI and machine learning from the perspective of the core problem that needs to be solved. Using image classification as an example, machine learning can determine if an image is a cat or not. Other machine learning examples like computer vision, identify where the cat is in the image and image captioning describes everything in the image. Best Match Recommendation, robotics and language translation are other AI applications used to solve best use cases for customers and businesses. Most of these problem cases identified above are well-defined and repeatable for machine learning programs. One direction in which AI is being developed is by finding the valuable, repetitive, but time consuming components of a business workflow and designing algorithms that can be applied to increase efficiency.