A Nature Inspired New Golden Age

A history of artificial intelligence

Optimism and ambition for artificial intelligence were flying high when the subject was formalized in the 1950s. Initial successes saw computers playing simple games and proving theorems. Some people were convinced that machines with human-level intelligence would appear within a decade or so.

But developing artificial intelligence proved hard, and progress stalled. In the 1970s, the academic challenges to creating artificial intelligence mounted and were followed by funding cuts and a loss of interest.

It seemed like machines with cold hard logic, using absolute 1s and 0s, would never be able to achieve the nuanced, organic, sometimes fuzzy thought processes of biological brains.

After a period of stagnation, an incredibly powerful idea emerged that lifted the search for machine intelligence out of its rut. Why not try to build artificial brains by copying how real biological brains worked? The idea was to mimic real brains with neurons instead of logic gates, and softer, more organic reasoning instead of black-and-white, absolutist, traditional algorithms.

Scientists were inspired by comparing the apparent simplicity of a bee’s or pigeon’s brain to the complex tasks they could do. Brains that weighed a fraction of a gram were able to steer flight and adapt to the wind, identify food and predators, and quickly decide whether to fight or attempt escape. Surely computers, with the new availability of cheap resources, could mimic and improve on these brains? A bee’s brain has around 950,000 neurons. Could today’s computers, with their gigabytes and terabytes of resources, outperform bees?

However, when using traditional approaches to solving problems, computers with massive storage and super-fast processors still couldn’t achieve what the relatively minuscule brains in birds and bees could.

Neural networks

The concept of neural networks emerged from this drive for biologically inspired intelligent computing, and has become one of the most powerful and useful methods in the field of artificial intelligence. Today, computers like Google’s DeepMind, which has achieved fantastic things like learning how to play video games by itself and beating a world master at the game of Go, have neural networks at their foundation. Neural networks are already at the heart of everyday technology, like automatic car number plate recognition and decoding handwritten postcodes on your handwritten letters.

In this course, we’ll discuss neural networks, learn how they work, and make our own neural network that can be trained to recognize handwritten characters—a task that is very difficult with traditional approaches to computing.