The Wonderful and Terrifying Implications of Computers That Can Learn

 Artificial Intelligence  

Machine learning is growing up. In a recent TEDx video, Jeremy Howard shows some of the advances in the technology, and asks us to consider what we'll have to do when it matures.

In  recent TEDx Talk, Jeremy Howard, CEO of Enlitic, talks about some of the potentials and threats of machine learning.

Howard covers some of the recent advances in deep learning, including translation, image recognition and even sentiment understanding. He covers work done at Google, and Baidu and others.

The same deep learning algorithm techniques that are being used to allow computers to write descripting sentences, are now being used increasingly in medicine.

Howard's Enlitic uses recent advances in machine learning to make medical diagnostics faster, more accurate, and more accessible. The company's mission is to provide the tools that allow physicians to fully utilize the vast stores of medical data collected today, regardless of what form they are in - such as medical images, doctors' notes, and structured lab tests.

Jeremy Howard
Image Source - Michael O'Donnell/VentureBeat

Related articles AI System Breaks CATCHA Turing Test  Facebook Researching AI In a Big Way  Robin Hanson Envisions a Future Robot Society

Enlitic aims to process huge heaps of CT scans, X-rays, and other kinds of images and then unearth latent patterns in new images. It will do that using a deep learning.

"This is not a case where the human is being replaced by a computer," Howard says. "What we are doing here is replacing something that used to take a team of about five or six people about seven years, and replacing it with something that takes about 15 minutes for one person."

Howard sees deep learning as a tool that can help extend medical knowledge.  Pointing to data from the World Economic Forum, which shows there is a lack of medical expertise throughout the world, especially in the developing world, Howard suggests AI may help raise the bar for these countries.

On a more pessimistic note, Howard points out that deep learning algorithms present a real threat to our service-industry based economies.  In the so-called developed world, service industries account for nearly 80% of employment.

Deep learning is allowing computers to see, read, speak and integrate knowledge, all tasks that are part of service jobs today, and point to a massive and unprecedented amount of technological unemployment.

"Human performance grows at this gradual rate.  But we now know have a system, deep learning, that we know grows in capability exponentially. We see the things around us and say, 'computers are pretty dumb,' but in five years, the computers will be off this chart."

"If you think this has all happened before, we've seen the results in the past when new things come along and [people] get replaced by new jobs, what are these new jobs going to be?" asks Howard. "Its very hard for us to estimate this, because human performance grows at this gradual rate.  But we now know have a system, deep learning, that we know grows in capability exponentially. We see the things around us and say, 'computers are pretty dumb,' but in five years, the computers will be off this chart."

Unlike the plateaus in productivity after the initial surge of the industrial revolution, Howard says that computers will not stop improving, as they follow the trends of Moore's Law.  The pace of this change means that education will not even help according to Howard's graphics.

Howard is a serial entrepreneur, business strategist, developer, and educator. He is also the youngest faculty member at Singularity University, where he teaches data science, and is a Young Global Leader with the World Economic Forum. He advised Khosla Ventures as their Data Strategist, identifying the biggest opportunities for investing in data driven startups, and helping their portfolio companies build data driven businesses. Previously he was the President and Chief Scientist of Kaggle, a community and competition platform for over 150,000 data scientists. Before working at Kaggle, he was the top ranked participant in data science competitions globally, in 2010 and 2011. He founded two successful Australian startups (the email provider FastMail, and the insurance pricing algorithm company Optimal Decisions Group), both of which grew internationally and were sold to large international companies.

He started his career in management consulting, working at the world’s most exclusive firms, including McKinsey & Co, and AT Kearney (becoming the youngest engagement manager world-wide, and building a new global practice in what is now called “Big Data”). He is also a keen student, for example developing a new system for learning Chinese, which he used to develop usable Chinese language skills in just one year. Jeremy has mentored and advised many startups, and is also an angel investor. He has contributed to a range of open source projects as a developer, and was a regular expert guest on Australia's most popular TV morning news program "Sunrise".



SOURCE  TEDx Talks

By 33rd Square

Embed