Every day brings another exciting story of how artificial intelligence is improving our lives and businesses. AI is already analyzing x-rays, powering the Internet of Things and recommending best next actions for sales and marketing teams. But for every AI success story, countless projects never make it out of the lab. That's because putting machine learning research into production and using it to offer real value to customers is often harder than developing a scientifically sound algorithm. Many companies I've encountered over the last several years have faced this challenge, which I refer to as "crossing the AI chasm."

IBM was awarded the most patents in the US in 2016. IBM's efforts to match and surpass the human brain with computing technology helped push the company to the top of the 2016 list of patent awards. The US Patent and Trademark Office granted IBM 8,088 patents for the year, more than 2,700 of them stemming from artificial intelligence and cognitive computing work, IBM and IFI Claims said Monday. Next on the list was Samsung with 5,518 patents, Canon with 3,665, Qualcomm with 2,897 and Google with 2,835. In total, the USPTO granted 304,126 patents in 2016, 10 percent more than the year before, IFI Claims said.

Opinions expressed by Forbes Contributors are their own. The author is a Forbes contributor. The opinions expressed are those of the writer. "Microsoft Monday" takes a look back at the past week of news related to Microsoft. This week, "Microsoft Monday" includes details about a red Xbox One controller being released tomorrow, the accidental release of a Forza Horizon 3 developer version, Minecraft surpassing 25 million copies sold for PCs and Macs, a few Windows 10 user interface tweaks, a connected vehicle deal signed with Renault-Nissan and more!

Update: Machine Learning is Fun! Part 3, Part 4, Part 5 and Part 6 are now available! Also, don't forget to check out Part 1. In Part 1, we said that Machine Learning is using generic algorithms to tell you something interesting about your data without writing any code specific to the problem you are solving. This time, we are going to see one of these generic algorithms do something really cool -- create video game levels that look like they were made by humans.

Over the past 2-3 years or so, I've heard this buzz word being tossed around a lot, and it's something that has definitely seized my curiosity recently. So if you are from the field of Computer Science, I'm sure you also must have come across this term at least once. Deep Learning along with Neural Networks is an area of active research nowadays. Deep Learning is a sub-branch of Machine Learning. Before we dive more into Deep Learning, let's begin with the broader field of Machine learning.

We've all got bills to pay, fiscal responsibilities to meet, retirements to save for -- and we need help. As tens of thousands of chatbots proliferate across Slack, Facebook, and other platforms, you hear naysayers claim that bots might be useful someday but that today they're little more than a gimmick. It's hard to feel that way after you learn about a handful of chatbots that have saved people more than $10 million. Next time you hear someone make that argument, point them to this list, because perhaps nothing is more useful than putting (or keeping) money in your pocket. After its first month, about 100,000 people have interacted with PennyCat, according to maker Leo Kangin.

SANTA CLARA, Calif., and Eindhoven, Netherlands, Jan. 3, 2017 – Intel has agreed to purchase a 15 percent ownership stake in HERE, a global provider of digital maps and location-based services, from HERE's current indirect shareholders: AUDI AG, BMW AG and Daimler AG. In conjunction with Intel's acquisition of a stake in HERE, the two companies also signed an agreement to collaborate on the research and development of a highly scalable proof-of-concept architecture that supports real-time updates of high definition (HD) maps for highly and fully automated driving. Additionally, the two companies plan to jointly explore strategic opportunities that result from enriching edge-computing devices with location data. "Cars are rapidly becoming some of the world's most intelligent, connected devices," said Brian Krzanich, Intel CEO. "We look forward to working with HERE and its automotive partners to deliver an important technology foundation for smart and connected cars of the future."

From one direction comes intense competitive pressure to innovate or to at least follow very, very quickly. From the precisely opposite direction comes the potentially existential threat of an app very publicly flopping or - even worse - being very publicly revealed to jeopardize the customer's well-being. Either way, you lose brand value in a social marketplace where brand is your primary currency. Auto-da-fé The car industry offers an excellent case in point. One day, carmakers are climbing all over each other to position themselves as au courant with Apple CarPlay and Google Auto.

Human-robot relationships are a running theme in pop culture, from the cylons of Battlestar Galactica to Spike Jonze's film Her and last year's hit show Westworld. But that kind of scenario might not be science fiction much longer. Romance between humans and machines is already nearing the realm of the possible. This year, the California company Abyss Creations plans to start selling a new generation of high-tech sex robots -- dolls that can actually speak and respond to touch. And according to artificial intelligence expert Dr. David Levy, in a few generations, we won't just be having sex with robots, we'll be marrying them.

It is well known that the global optimum of a MDP with finite state and action sets can be obtained through methods based on dynamic programming. Unfortunately, these techniques are known to suffer from the curse of dimensionality, which makes them infeasible for many real-world problems of interest. As a result, most research in the reinforcement learning and control theory literature has focused on obtaining approximate or locally optimal solutions. There exists a broad spectrum of such techniques, including approximate dynamic programming methods, tree search methods, local trajectory-optimization techniques, such as differential dynamic programming and iLQG, and policy search methods. In this talk I shall provide an introduction to policy search methods, which are a family of algorithms that have proven extremely popular in recent years, and which have numerous desirable properties that make them attractive in practice.