Where did Neural Networks begin?

As we dive into the week, we'd like to take a moment to reflect on the remarkable journey that artificial intelligence (AI) has taken up until now.

AI 's Conceptual Beginnings

While many associate AI with recent advancements, it's important to acknowledge its roots, which extend deep into the 20th century. The conceptual beginnings of AI can be traced back to the brilliant mind of Alan Turing. Turing, renowned for his cryptographic work during World War II, posed a pivotal question: "Can machines think?" However, he believed that framing the issue this way was not ideal. Instead, he introduced the "Imitation Game," a thought experiment designed to determine if a machine could convincingly imitate a human through text-based communication. This seminal idea laid the foundation for the AI journey we're still on today.

Kick-Starting the Idea of Neural Networks

Frank Rosenblatt in 1958, took a significant step forward by introducing the "Perceptron," a computer modeled after the brain's neural structure. This machine could teach itself new skills by making predictions and adjusting based on its errors, mimicking how neurons learn. Unfortunately, Rosenblatt's concept faced limitations, as it relied on a single-layer neural network, proving inadequate for meaningful machine learning. Despite this setback, Rosenblatt's work was a critical milestone in the AI journey.

Many Layers Make Machine Learning Work

The breakthrough came in the 1980s when Geoffrey Hinton, alongside researchers like Yann LeCun and Yoshua Bengio, proposed that multiple layers and an extensive network of connections were essential for AI success. This insight mirrored the neural network structure found in the human brain. Neurons in your brain take data and produce a corresponding output based on their training, using electricity and chemicals to send messages to each other. AI neural networks use input like math and data to learn, just like our brain learns from experiences, which allows them to recognize patterns, make decisions and solve problems.

This is part of the reason we chose Qortex as a name, as in the brain's cortex or the cerebral cortex.??

The cerebral cortex is made up of a vast network of neurons (nerve cells) that are densely packed together. These neurons form intricate connections with each other, creating a complex neural network. This network enables the brain to process information, make decisions, and carry out various cognitive functions. Very much like how Qortex's proprietary AI?analyzes videos, identifies key moments, categorizes them, and strategically serves advertisements during the most relevant and engaging segments to enhance user experience.

AI Today

However, it wasn't until the 2010s, with advancements in hardware and data sets, that AI truly took off. Moore's Law played a pivotal role, enabling researchers to train neural networks effectively. This marked the era where AI started making significant impacts in various domains, from smart assistants to autonomous vehicles.

As we fast-forward to late 2022, companies like Qortex and of course, ChatGPT's emergence brought AI into the mainstream spotlight, showcasing its incredible capabilities. Yet, we are only beginning to scratch the surface of what AI can achieve. The future of AI remains an open frontier, full of possibilities that we can only imagine.