imagen de una persona con un teclado, está tecleando

The Brain behind Artificial Intelligence

fathooo
Article, Artificial intelligence, Psychology

An Approach to Artificial Intelligence.

There are many theories about how AI (Artificial Intelligence) works in popular knowledge. The general functioning and concrete notion is that a machine is capable of learning or behaving like a human. However, the learning processes vary depending on the considerations taken into account when "teaching" a machine. In this article on my blog, I will take psychology into account as an initial step to understand some elements, and it can also provide us with the necessary concepts to understand the functioning of AI. The psychological mechanisms behind the functioning of AI could provide us with information on how to train it better and help us potentially guide us in the design of better machine learning systems.

What is Artificial Intelligence?

Artificial Intelligence is the ability of a machine to learn and behave like a human. It has characteristics such as reasoning, learning, creativity, and the ability to plan. There are fields of computer science that study and develop systems that can perform tasks that require human intelligence, such as pattern recognition, problem solving, driving vehicles, fraud detection in banking, etc. With the various sensors, devices, and technologies available today, technological development has advanced tremendously in this area in recent years. Although the concept dates back to 1956 when a group of scientists initiated the "Artificial Intelligence" research project at Dartmouth College in the United States (Salesforce BLOG, 2017).

Four Approaches to Artificial Intelligence:

Depending on the type of Artificial Intelligence, its approach to understanding is different, as those who design it take different approaches to achieve the goal of creating a machine capable of learning or behaving like a human. While there are various conceptions regarding the topic, the classification by Stuart J. Russell and Peter Norvig will be presented below (Referencias), the complete book can be found in the references section, and I also provide more interesting material for you to delve deeper into the topic.

  1. Systems that think like humans: This approach strives to create systems that think like humans, that is, that can reason and plan. These are machines that are linked to human thought processes, such as decision-making, problem-solving, planning, etc. Neural networks are an example of this type of system.

  2. Systems that think rationally: These are systems that involve a combination of mathematics and engineering using computational models. The aim is to emulate reasoning and investigate how machines can perceive, reason, and make decisions.

  3. Systems that act like humans: The effort to develop machines capable of performing tasks that require human intelligence, such as pattern recognition, learning, problem-solving, etc. An example of this type of system is robots.

  4. Systems that act rationally: These are systems related to intelligent behavior. Similarly, the goal is to emulate human reasoning at the level of behavior, that is, to enable machines to act rationally.

Image of a person with a keyboard, typing

How does the human brain work?

The human brain is a complex organ that processes the information it receives from the environment and responds to the organism's needs. It controls vital functions such as body movement, sensory perception, and pain perception. Nerve fibers stretch between the brain and the spinal cord, transmitting signals between the brain and spinal cord and vice versa.

While the human brain is a complex organ, it can be divided into three main parts: the cerebrum, the cerebellum, and the brainstem. The cerebrum is the largest part of the brain, responsible for processing the information it receives from the environment and responding to the organism's needs. The cerebellum controls body movement, sensory perception, and pain perception. The brainstem processes the information it receives from the brain and spinal cord and vice versa.

Similarities between the human brain and machines:

Both the conscious and unconscious processes of the cortex can be thought of as computational processes of varying priority controlled by different sets of computers with different functions. Sensory information, previously processed by the Front End Processors (FEP), reaches the three major processing nuclei. In one of them, an image of reality is generated, in another, information is temporarily stored in a "cache" (the amygdala), while new concepts are defined and stored (the hippocampus) (Martínez y Fornaguera, 1998). Although the above analogy is not far from reality, as the human brain is capable of processing information in a similar way to a computer, the human brain can be considered an information processing machine.

Surprising Things about AI... It's Just the Beginning.

It is now common to see in the news that software has been developed that can recognize objects or can speak and respond like a human. It is also common to see in Instagram or TikTok reels the ability to request a software through text and have it generate an image or an article from scratch. In the field of health, algorithms have already been created that can detect diseases such as breast cancer or detect abnormalities in the heart, etc. In the case of Alzheimer's, the biggest challenge is to achieve early diagnosis and slow its progression. In that sense, a team led by Marianna La Rocca, a physicist at the University of Bari in Italy, developed a machine learning algorithm capable of discerning structural changes in the brain caused by Alzheimer's (BBVA OPENMIND, 2018). There are more and more applications that can be performed with AI, and more and more companies are joining this trend, as AI is capable of performing tasks that were previously impossible.

Neuronal system, neural networks in the form of lights

In the coming decades, with improved and more interconnected computer devices, even more noticeable paradigm shifts are expected in the relationship between AI and human beings. The computing and data capabilities are increasing, as well as the learning capabilities of AI. In the future, AI may be able to perform tasks that were previously impossible to carry out, such as creating art, generating music, writing texts, etc. Overall, many studies support the interesting possibility of quantifiable similarities between the brain and AI models, which can help us better understand the functioning of the human brain.

References