Practical, Affordable, Artificial Intelligence — Build a Deep Learning Organization
How you can leverage AI in a meaningful, affordable, and understandable way.
Warning: this article was produced without the expressed written consent of our soon-to-be master robot overlords. I tried, but there didn’t seem to be any around to ask.
Artificial Intelligence is all the buzz. As a result it is overhyped, overpriced, and (more intriguingly) overlooked. Many viewers will see AI in this title and zoom right by. They don’t understand it, aren’t sure if it is real, but are positive they can’t afford it. And they aren’t wrong…
At the highest levels, Artificial Intelligence is the playground of Google, Facebook, the NSA, and NASA. If you are managing a multi-billion dollar stock portfolio, you likely have a team of quants on the payroll. Big banks and big pharma have huge analytic divisions that dabble here as well. And of course, Universities like Carnegie Mellon have staked their reputation on Machine Learning.
None of this is very helpful to the average start-up (outside the AI space), the pre-IPO, or even most mid-to-large size companies. Most take the roll of disinterested bystander or skeptical observer.
A wealth of new start-ups are pushing to bring Artificial Intelligence to more people, but if you are not in cybersecurity, robotics, or healthcare — accessibility feels years away (perhaps decades).
New technologies/languages/frameworks like Tensor Flow, Torch, Teano, and Spark are being learned by a growing number of enthusiastic 20-somethings. But while enthusiastic, these young professionals are a decade away from the business insight needed to make AI an efficient, effective, or even tangible reality for most businesses. It hasn’t kept a few businesses from trying…
To Make Artificial Intelligence Accessible — You Need To Look Deeper
At the bottom of our cover art hierarchy and still living in relative buzz-obscurity is the concept of Deep Learning. I have written about it at length, both the history and general explanations. Essentially, Deep Learning creates a framework consisting of alternating layers of human “supervision” and machine learning optimization. It is a bit like building a club sandwich.
Using that analogy, think of the human supervision as the bread. This could be a whole article in itself (coming soon) but bread works on many levels. It is the most common ingredient in any club sandwich, it is cheap, easy to make, and the best thing since… well you get the point. Or at least you will.
You see, that structured layer of the club sandwich is what creates the great experience. Sure the other ingredients matter, but club sandwiches are built on structure. Your business can benefit from this, too. In other words, the concepts behind deep learning can be built into your business. By leveraging the insight and learning freely available from this newly popular field, you can organize and engineer your business to make better use of learning and feedback.
It is important to remember that in deep learning, artificial intelligence is leveraged for equal parts insight development and optimization. The latter is a luxury to most companies anyway and the former can be performed by knowledgeable members of your staff. The key is that the structure (the bread) empowers your organization to do this more effectively and efficiently.
Down the road and as budget permits, you can look to invest in select and limited (read affordable) opportunities to augment your staff with smaller amounts of automation, machine learning, and data science. This may take a while, but in the meantime — this structure alone will be generating meaningful and incremental returns for your company.
How Will Committing To A Deep Learning Structure Change Your Organization?
This article can only go so far, but more will follow (coming soon). We can focus now on a few of the major changes. Deep Learning organizations -
- Make data accessible to everyone
- Focus on feedback over data
- Define clear layers of accountability
- Make learning a priority
- Invest in transparency
- Measure, track, and report
- Predict and imagine
Deep Learning will change your business culture. Whether you intend to or not, these changes begin powerful feedback loops in your organization. That is the very core of any deep learning organization. They handle change because they are constantly changing and evolving — only with the appropriate structure, they often don’t notice. It is just business as usual.
Deep Learning drives communication. It makes a communication plan irrelevant, effectively replacing it. It mitigates the need for endless meetings. People understand the needs and goals of the organization because they are an ever present part of their jobs. It keeps your business young.
If this seems too good to be true, note that most start-ups don’t worry about communications plans or a malaise of meetings — the smallest companies have deep learning structures organically. Unfortunately, without a strong understanding of why — growth quickly breaks those early structures and quickly learning organizations turn into deep bureaucracies.
Getting started requires only a few simple steps… a map, a story, a process, and some simple training can go a long way. Deep Learning scales well. It gets more complicated with time, but early steps will assure the progression is easier, cheaper, and more effective. So start early. And keep reading… more articles coming soon.
For help getting your business started-
