2018 is going to be a year of major changes in every profession, and it’s just a matter of time before you begin to hear about “deep learning.” It’s likely that you haven’t heard of it yet, but it’s impossible that you’ve gone without being affected by it.
Deep learning is a technique a programmer can use to make machines develop their own algorithms and refine them as they progress, often with the purpose to identify trends in data. It’s been used in recent times to make machines capable of recognizing things such as objects in an image or video, or speech in audio to generate subtitles. In the local area, we can see developments in industries such as pest control, where insect infestations can be predicted before they ever happen or car dealerships offering auto loans when machines identify potential customers (Although, such implementations are too pricey today as things are.). With deep learning, Stanford researchers created an AI that could assist in researching new drugs by accurately predicting toxicity and side effects of a proposed drug before researchers started physically testing the drugs on living subjects. Most things in the world can be predicted with enough data, it’s just a matter of spotting the trends, and with deep learning, that is increasingly getting easier to accomplish every day.
Google most famously uses deep learning AI for basically everything. Google first announced their use of deep learning in the “Google Brain” project in 2011. Since then, deep learning now handles Google translate, it offers YouTube video suggestions, it is crawling the images of the Internet for more accurate image searches, and I would be willing to bet that the text search results implement it as well.
Soon, it’s likely that we will see stocks exclusively being traded by deep learning AI, while ad companies will begin to transition into digital data farms. In a world where AI can figure out any trend in the data, the bottom line will entirely depend on how much data you have, and how quickly you can process it. And while this sounds exciting, there is also a great deal of concern generated because of the current landscape on the cutting edge.
With research being done in the US, we are currently the leading nation in this new technology. However, as we cut deep learning funding to the National Science Foundation, China is doubling down on their deep learning investments. With many jobs associated with deep learning still being in the research stages, research funding overseas is forcing many of our graduates to move to China to pursue developing these new technologies with shrinking opportunities here at home. As deep learning becomes a necessary technology in the business world, we risk losing to China in every field if we don’t find the funds for deep learning research.
Though there is a reason to be concerned, we certainly aren’t out of the running yet for the AI arms race with China. It’s still at the bleeding edge of technology, and as there’s less than a decade of history in this race, it’s only just begun. When the world starts to change, and you see these algorithms begin to evolve how we do business, deep learning is going to be the word you’re looking for when you decide to explain it to your friends and co-workers.