Neural Networks

Neural Networks are a form of Artificial Intelligence that learns through a process called Machine Learning.

Neural Networks originated in the 1970’s when scientists tried to recreate the logic of a brain using virtual neurons.  A neuron works by taking input from a variety of sources, mainly other neurons, and when a threshold is reached, it fires. One neuron will often feed into many other neurons creating a complex network of interactions. It is this process of input, threshold and firing that is the foundation of thought, memory, and you.

Brain impulses. Thinking prosess.

In terms of Thinking Exponentially, neural networks scale up with computing power and training set size.  As computing grows, becoming faster and smaller, so do the capabilities of Neural Networks. It will only take a few short years for Neural Networks to match and exceed human intelligence. I expect to see this happen in the 2020’s. Between now and then you will see Neural Networks drive robotics, artificial intelligence and automation.

IBM has an especially promising project called IBM Cognitive Computing, which effectively is trying to build and scale neural chips. By building the neural logic in configurable hardware, this approach will scale extremely well.

An interesting demonstration of adaptive neural networks is found in this video. Watch closely on how it learns and develops. That is the key to true intelligence.

For comparison, the human brain consists of about 100 billion neurons, each connected to each other with over 1000 connections apiece. That means there are roughly 100 Billion * 1000 connections, or 100,000,000,000,000 connections in your 3.5 pound brain – and you can run it by eating an apple.

Today the best neural networks are less than 1% of the human brain, take up an entire floor, and need a small nuclear reactor to power. If you Think Exponentially, you realize that it only takes 6-8 generations in computing to create the power required for human level machine intelligence.


Categories: AI, Video

Author:Geoffrey Shmigelsky

Technology Entrepreneur, Angel Investor


Subscribe to our RSS feed and social profiles to receive updates.


  1. Artificial Intelligence | Think Exponential - February 14, 2013

    […] are many kinds forms of artificial intelligence: Neural Networks, Genetic algorithms, or  Expert systems. These are called Weak AI as they can only do one or two […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: