“A place I can imagine machine learning working for an artist would be not in developing an independent subjectivity, like writing a poem, but actually filling in gaps that are to do with labor, like the way Photoshop works with different tools that you can use.”
Rachel Rose
We, humans, have developed all sorts of tools for thousands of years to make our lives better.
Constantly looking for ways to augment our work is what mostly separates us from the rest of the animal kingdom.
Over time, these tools have become more complex and more helpful. These tools have elevated billions of humans out of poverty. Without these tools, the industrial revolution would not have been possible and the world that we know today would not exist.
What’s being creative?
Creativity is the last great mystery. What is creativity? How does a new idea pop into our minds?
Creativity is all about problem solving:
- Find a problem: The problem must be one that opens a new dimension of questioning. Seeing a problem that no one knew was there.
- Spot connections: Come up with assumptions of connections between areas which at first sight have nothing to do with one another.
Creativity helps us come up with products, ideas, or artifacts that are new, useful, and surprising. Creativity is mostly a reinterpretation of existing solutions.
Creativity can only be measured in subjective terms. Some of the most important creative ideas have not been recognized as such for years.
Types of creativity
In her seminal work “The Creative Mind” Margaret Boden explains the three types of creativity:
- Combinatorial creativity: Combining familiar ideas in new ways. For example, comparing a heart to a pump or the atom acting as a minuscule solar system. You combine different words to create your own poem.
- Exploratory creativity: Pushing the boundaries of the familiar. Working within the expected rules of a procedure as in arts, literature, music, and science and generating new styles, new sentences, new melodies, and new theories. It’s like exploring a familiar place and finding something new there.
- Transformational creativity: The great creative leaps of the imagination. You work within the constraints and then you burst out. You end up rewriting the rule book. Like how Einstein discovered special relativity.
The 4 C model of creativity:
- Mini C: The child who builds their first sand castle is at their mini-C level of creativity.
- Little C: The creator has received feedback on previous work and has tried to create something of value.
- Pro C: The creator has put significant study or works in a field. This person has professional proficiency in a given area. The creator’s contributions might not be known to the general public.
- Big C: These are the creators who come to mind when we think of creative people. They are icons in particular fields and are immortalized in museums and history. Like Einstein with his relativity theory.
Stages of creativity
Creativity begins with an intention to solve a problem. You can start by imagining different ways of solving a problem. Then, go into a dream-like state; beyond logic and standard knowledge.
After you mull over different approaches to solve a problem, the unconscious mind takes over.
Unconscious thought primed by conscious thought is the key to the creative process. It's what causes the moment of illumination.
The scientist starts with deliberate thought and then drifts into a wandering state. And refines the idea with deliberate thought once an insight is received.
That eureka moment like Archimedes in the bath. When we suddenly see the solution to a knotty problem with blinding clarity. It's the creation of something from nothing. Like a novelist dreaming up a story or a poet dreaming up a poem, a melody that nobody has thought of before.
- Conscious thought: Ideas don’t arise in a vacuum. Feed your brain with the material you’re working on. Sow the seeds.
- Unconscious thought: Take a break. The passionate desire to solve a problem keeps it alive in the subconscious. Wait for your seeds to germinate. Let the mind wander.
- Illumination: Wait for the “Aha” moment. Just as you stop thinking of a problem, the solution flashes through your mind.
- Verification: Use critical thinking skills to fine tune your new found idea.
What can you do to be more creative?
Here’s how you can nurture your creativity:
- Introspect: Sit in silence without any distractions. Mindfulness and meditation helps you be more creative.
- Take time off: Stop conscious work on a problem so that your unconscious can start its work of reflection. Go for a walk, take a nap, and the solution might just come to you out of nowhere.
- Hone your skills: Determine the one best thing that you’re good at and pursue that.
- Make mistakes: Make attempts that fail quickly so that you can move on to new ones right away.
- Collaborate: Closely collaborate with people. Learn from the work of others.
- Borrow great ideas: Use the ideas of others to stimulate your own fertile mind.
- Have faith: Have faith in yourself and your instincts.
To be creative, you have to build up your expertise by accumulating a vast amount of knowledge of a subject. Constantly sifting through this background knowledge helps you find creative solutions.
AI, getting creative
The pursuit of being creative applies equally to the brain as well as to the computer.
As AI gets more sophisticated, one day it might be creative like us. In principle, because the brain obeys the laws of physics, the computers can do anything the brain can do.
Thinking consists of receiving perceptions that the brain acts on and uses to create new knowledge. The computer is fed data, processes the data, and uses it for problem solving.
Any traits specific to humans—imagination, creativity, even consciousness may just be equivalent to software programs.
Could AI spot connections between disparate ideas? Can AI have flashes of inspiration like we do? Could they say something that nobody thought of saying ever before? Could they dream up plays like that of Shakespeare? Or, maybe they would come up with solutions to problems but just not in the same way that we would come up with.
Imagine a world in which AI produces new works of art and music, discovers scientific theories, writes stories, makes business decisions, and tells jokes.
Emotions, spurring creativity
Will AI ever experience suffering–one of the hallmarks that hones creativity? To do so, it would need to have emotions. Perhaps one day there will be robots with complex systems of sensors, regulatory mechanisms, and communication pathways that duplicate human emotions. Later, they might even go beyond humans and develop new and even unimaginable feelings.
In the future, it might be possible for AI to miss someone and experience grief when they realize that the person is no longer available.
Giving the AI emotions might affect the complexity of its hidden layers and open the possibility of exhibiting unpredictable behavior that's an essential element in creativity. Thus, it might, on its own, decide to do something new. Maybe AlphaGo gets spurred by the boredom of playing with itself and invent a new game.
Evaluating creativity in AI
How do we assess the degree of creativity an AI model possesses? Maybe, the best way is by using human judgment.
Other aspects might include, the lesser the amount of data an AI model trains on, the greater is its creativity. The extent that the final output differs from the training data. How fin-tuned the model is and so on.
Greame Ritchie has drawn up a list of highly mathematical criteria framed in equations to measure the level of creativity of any AI model - Some Empirical Criteria for Attributing Creativity to a Computer Program.
Anna Jordanous gets at computer creativity by defining the components of creativity and ranking them in sequence - A Standardised Procedure for Evaluating Creative Systems.
According to her study, creative work doesn't even need to be good. Because the 'good' a work is ranks close to the bottom of the list.
Conclusion
Creativity is not limited to people. Humans are biological machines. If humans can create art, why not machines?
The word Eureka roughly translates to ‘I have found it’ in Greek. It was the words used by Archimedes when he discovered how to measure volume from a regular shape. As he stepped into a bathtub he noticed that the water in the tub rose because of his added volume. At that moment, he realized that the volume of the body must be correlated to the displacement of water. He was so excited with this insight that he ran through the street naked to share this discovery.
It’s hard to imagine a computer having a Eureka moment, because for that, it needs to be able to tap into its subconscious while engaging in undirected mind wandering.
The intuitive exchange between the conscious and subconscious is essential for true creativity.