Machine Learning | Are Computers Awake?
Over the years in my Machine Learning journey, I constantly find myself ruminating over the concept of a learning machine.
What is Learning? What does it mean for a computer to learn?
Learning itself is an integrated part of human existence.
Some scientists believe that humans accord themselves the rank of a higher species not just because of our ability to learn and adapt to our environment, but our higher consciousness.
A self-reflective consciousness.
“Here is this three-pound mass of jelly you can hold in the palm of your hand…it can contemplate the meaning of infinity, and it can contemplate itself contemplating the meaning of infinity.” Neuroscientist Vilayanur Subramanian Ramachandran
“Learning is the acquisition of knowledge or skills through study, experience, or being taught.” Oxford Dictionary
We are not the only species that can learn. Animals and even plants can acquire knowledge and adapt to their environment through learning.
Computers are designed to be discrete. It takes an input and responds with an expected output. It can also store information.
What does it mean for a computer to learn?
A computer can store and retrieve information. Does it mean that it is learning?
To answer this question, we have to look into the species that the concept of a “learning machine” is modeled after. Yes you guessed right, Humans 😃
For example, I signed up for a driving class to learn how to drive a car.
I know at the end of my training, I will be expected to take a test before I can be certified to drive with a license.
This means that I will be expected to have learned how to drive.
In training, my instructor and I drove through the training arena. She showed me how to steer the car, accelerate and stop the car.
After few weeks of training, I believed that I was conversant with the concept of driving. I know the road rules and my 9–3 hand position on the steering.
To prove that I have learned how to drive, I have to take a test.
Unlike my training arena with safety reflective cones and yellow tapes, I have to face a busy street filled with jaywalkers, traffic, and other road users (including ducks).
If I apply the concepts that I acquired during my training to successfully navigate this new terrain, then it is certain that I have truly learned how to drive.
The answer to the question above is not yet. A computer storing and retrieving information is not proof of learning.
But what about Artificial Neural Networks?
The subject of human consciousness is also complex. How can we create machines to be the very thing that we don’t fully understand yet?
How does a computer learn?
Model after humans, they also undergo training and testing. Data is fed into an algorithm to produce a learning model.
Machine learning algorithms are evaluated in an unseen space just like my driving live test.
If you want to learn more about Machine Learning, check out my blog tutorials
Does a computer know that it is learning?
Humans are conscious beings and self-aware. A computer will need to be self-aware to know that it is learning. For now, computers are not conscious. It is still a tool counting its 0s and 1s.
I’d love to hear your thoughts, please leave me a comment below
Connect with me on Twitter
