Looking ahead, we face some difficult – and yet exciting – research challenges in AI design. AI systems feed off of both positive and negative interactions with people. In that sense, the challenges are just as much social as they are technical. We will do everything possible to limit technical exploits but also know we cannot fully predict all possible human interactive misuses without learning from mistakes. To do AI right, one needs to iterate with many people and often in public forums.

IBM researchers, Tayfun Gokmen and Yurii Vlasov, unveiled a paper in which they invented the concept for a new chip called a Resistive Processing Unit (RPU) that can accelerate Deep Neural Networks training by up to 30,000x compared to conventional CPUs. A Deep Neural Network (DNN) is an artificial neural network with multiple hidden layers that can be trained in an unsupervised or supervised way, resulting in machine learning (or artificial intelligence) that can "learn" on its own. This is similar to what Google's AlphaGo AI has been using to learn playing Go. AlphaGo used a combination of a search-tree algorithm and two deep neural networks with multiple layers of millions of neuron-like connections. One, called the "policy network," would calculate which move has the highest chance of helping the AI win the game, and another one, called the "value network," would estimate how far it needs to predict the outcome of a move before it has a high enough chance to win in a localized battle. Many machine learning researchers have begun focusing on deep neural networks because of their promising potential.

IBM researchers, Tayfun Gokmen and Yurii Vlasov, unveiled a paper in which they invented the concept for a new chip called a Resistive Processing Unit (RPU) that can accelerate Deep Neural Networks training by up to 30,000x compared to conventional CPUs. A Deep Neural Network (DNN) is an artificial neural network with multiple hidden layers that can be trained in an unsupervised or supervised way, resulting in machine learning (or artificial intelligence) that can "learn" on its own. This is similar to what Google's AlphaGo AI has been using to learn playing Go. AlphaGo used a combination of a search-tree algorithm and two deep neural networks with multiple layers of millions of neuron-like connections. One, called the "policy network," would calculate which move has the highest chance of helping the AI win the game, and another one, called the "value network," would estimate how far it needs to predict the outcome of a move before it has a high enough chance to win in a localized battle. Many machine learning researchers have begun focusing on deep neural networks because of their promising potential.

Everyone is looking for the next breakthrough in machine translation. No one believes that machine translation is a completely solved problem. Most people would like to see machine translation systems produce higher quality results. A good translation is one where the meaning of the source is preserved, and it is rendered correctly in the target language. Users expect accuracy on all of the various levels–grammar, syntax, semantics and pragmatics.

News of Google's self-driving car getting in a minor collision with a bus has been all over the internet recently. At the same time, I also came across this article in Forbes magazine, "Machine Learning Needs a Human-In-The-Loop." Both topics raise questions about the boundaries of autonomous operations.I started to consider this in relation to my own experience in the world of IT automation. The fact is that while automation is now deeply embedded in most manufacturing processes, IT has been comparatively slow in routinely applying automation technology to large areas of IT and security operations. Why is that the case?

Over the past couple of years I have had the good fortune to discuss the need for improved understand of profitability and capital utilization in broker dealers. Discussions on the need to find the right clients to meet the balance sheet utilization mix meander with no clear end in sight. Investment managers are moving their books of business, not based upon desire, but based upon the BD needing to achieve the correct balance sheet mix. BDs are looking at return on capital as much as profitability of clients. Machine learning is a field of artificial intelligence that involves developing self-learning algorithms.

In my previous post "The Human Face of Big Data" (attached again below), I referenced a couple of blog posts from David Meyer, Chief Scientist and Brocade Fellow. Below is a great video of him presenting on Machine Learning and emerging applications for computer networking at the DevOps Networking Forum a couple weeks ago. Definitely worth the 30 minutes to watch if you're at all interested in the most current developments in this space. If you're interested in hearing David speak live, he'll be discussing this topic with regard to the Federal space at our upcoming Federal Forum conference in Washington, DC at the Renaissance Hotel on June 14. Register here and use the Promo Code: brocadefriends to attend for free.

AI as management assistant: The artificial intelligence program AlphaGo got a lot of attention for beating 18-time Go world champion Lee Sedol four out of five games last week. The significance of this achievement is rooted in the extraordinary number of possible moves in Go: 2.08168199382 … 10170, reportedly more than the number of atoms in the universe. That's too many possibilities for brute computing force to handle (which is how IBM's Deep Blue beat chess master Garry Kasparov 20 years ago). Yet AlphaGo, created by Google DeepMind, formerly British AI company DeepMind Technologies, mastered the 2,500-year-old board game on its own in a matter of months. "It started by studying a database of about 100,000 human matches, and then continued by playing against itself millions of times," reported science correspondent Geoff Brumfiel at NPR. Go bragging rights are nice for Google, but what does AlphaGo's victory mean for management?

On March 23, Microsoft launched an artificial intelligent (AI) chat bot on twitter called TayTweets, and within hours, anonymous individuals had hijacked the experiment. Because the program was not built with contingencies meant to handle inappropriate questions or information gathering, like a human brain for instance, by asking the AI bot something in a specific manner people were able to get responses regarding Holocaust denial, 9/11 conspiracies and incest. On their project website, the Microsoft development team stated the bot was designed to conduct research on "conversational understanding" through engaging and entertaining people with playful conversation. "Unfortunately, within the first 24 hours of coming online, we became aware of a coordinated effort by some users to abuse Tay's commenting skills to have Tay respond in inappropriate ways. As a result, we have taken Tay offline and are making adjustments."

Every week, CIO Journal offers a glimpse into the mind of the CEO, whose view of technology is shaped by stories in management journals, General interest magazines and, of course, in-flight publications. AI may undermine big-company advantages. Machine learning – software that can improve itself without human intervention – may mean trouble for big companies that depend on their heft to outmaneuver smaller upstarts, writes Howard Yu for the Harvard Business Review. And for a sneak preview of where the world is headed, one need not look further than the success story of AlphaGo, an artificial intelligence that beat a champion of the ancient game of Go, something that was previously thought to be impossible. "It is easy to imagine a world where self-taught algorithms will play a much bigger role in coordinating economic transactions; AlphaGo simply shows us what is possible in the near future. With instantaneous adjustment, automatic optimization, and continuous improvement all quietly managed by unsupervised algorithms, the redundancy of production facilities and wastage in the supply chain should become headaches of the past."