The puck-shaped Nexus Player is no longer available for sale from Google, as The Verge first reported. Other retailers haven't stocked the device for months. If you count the ill-fated Nexus Q, the Nexus Player was Google's third run at creating a set-top box, following a run of over-priced devices powered by the now defunct Google TV. The Nexus Player was first introduced in late 2014 as one of the first devices running Android TV. At the time, we said the player was a "fine first draft."

An Amazon Prime package awaits pickup by its customer. SAN FRANCISCO -- Amazon says it hasn't stopped price matching -- because it never had such a policy in the first place. Over the last several days, news stories and online forums reported the retail giant had ended its policy of reimbursing customers if the price of an item on its site drops within seven days of delivery. Amazon (AMZN) spokeswoman Julie Law said the company does not do what's known as "price matching" or "price protection," though its customer service associates are empowered to make decisions on behalf of the customer when it seems appropriate. "We've always had a no price matching policy, because we believe we're always making the best pricing decisions on behalf of our customers," she said.

An Amazon Prime package awaits pickup by its customer. SAN FRANCISCO -- Despite multiple news stories over the past several days that Amazon has ended its policy of reimbursing customers if the price of an item on its site drops within seven days of delivery, the company says it hasn't -- because it never had such a policy in the first place. Amazon (AMZN) does not do what's known as "price matching" or "price protection," though its customer service associates are empowered to make decisions on behalf of the customer when it seems appropriate, said spokeswoman Julie Law. "We've always had a no price matching policy, because we believe we're always making the best pricing decisions on behalf of our customers," she said. Because there's no actual policy and because customer service associates have broad freedom in what they offer customers, it's possible customers have in the past, and still are at times, getting price matching, Law said.

Thanks to machine learning, the page you see when you log-on to Amazon.com is likely very different from the one I see. Advertising, product recommendations, and special deals are all tailored to our unique customer profiles based on historical browsing trends and buying behavior. Online retailers like Amazon were among the first users of customer data collection and analysis for improving services and personalizing the shopping experience, and they've become so skilled some sites might even be able to predict what we will purchase before we even know what we're looking for. Advancements in digital technologies have driven a paradigm shift in the way businesses interact with their customers, with touchpoints increasingly moving to digital mediums. Because of the limited opportunities to satisfy customers on a person-to-person level, machine learning is now in widespread use by a variety of modern enterprises as a way to enrich customer experiences, create more personalized and customer-centric interactions, and offer seamless omnichannel communications. Machine learning goes a step beyond Big Data analytics, where machines employ advanced algorithms to autonomously adapt and learn from previous experiences, and therefore emulate the thought process behind human decision-making.

Chatbots already have a little bit of a bad name. Early reviews for the ones on Facebook Messenger have been rough due to apparent malfunctions, and Microsoft's Tay has been an utter disaster, at least on a couple of occasions. But a startup called ReplyYes, which offers a text-to-buy system for retailers, provides a glimpse into the potential of automated messaging. Interestingly, the company has a pair of e-commerce ventures. One sells vinyl records, the other graphic novels.

At the Stanford shopping center in Palo Alto, California, there is a new sheriff in town – and it's an egg-shaped robot. Outside Tiffany & Co, an unfortunate man holding a baby finds himself in the robot's path. It bears down on him, a little jerkily, like a giant Roomba. The man dodges but the robot's software is already trying to avoid him, so they end up on a collision course. "I've seen Terminator," the man says, half to himself and half to the amused crowd, "and that is some Skynet-ass shit."

Grace Peters recently wrote in HPCwire, "Thanks to machine learning, the page you see when you log-on to Amazon.com is likely very different from the one I see. Advertising, product recommendations, and special deals are all tailored to our unique customer profiles based on historical browsing trends and buying behavior. Online retailers like Amazon were among the first users of customer data collection and analysis for improving services and personalizing the shopping experience, and they've become so skilled some sites might even be able to predict what we will purchase before we even know what we're looking for." Peters goes on, "Advancements in digital technologies have driven a paradigm shift in the way businesses interact with their customers, with touchpoints increasingly moving to digital mediums. Because of the limited opportunities to satisfy customers on a person-to-person level, machine learning is now in widespread use by a variety of modern enterprises as a way to enrich customer experiences, create more personalized and customer-centric interactions, and offer seamless omnichannel communications."

Let's check in with who's on the AI (artificial intelligence) train. Apple kicked things off this year, announcing back in January that it had purchased Emotient, an AI tech startup in the facial recognition business. Did Apple say why it bought that company? Of course, it did not; mind your own business. Jump ahead to spring, and the next big company in the retail space (specifically, the biggest company in eCommerce) to make an AI-related move this year was Amazon.

A total of 2,552 players on over 2,000 teams participated in the Home Depot Product Search Relevance competition which ran on Kaggle from January to April 2016. Kagglers were challenged to predict the relevance between pairs of real customer queries and products. In this interview, the first place team describes their winning approach and how computing query centroids helped their solution overcome misspelled and ambiguous search terms. Andreas: I have a PhD in Wireless Network Optimization using statistical and machine learning techniques. I worked for 3.5 years as Senior Data Scientist at AGT International applying machine learning in different types of problems (remote sensing, data fusion, anomaly detection) and I hold an IEEE Certificate of Appreciation for winning first place in a prestigious IEEE contest.

Major corporations use this kind of artificial intelligence to help with the complexities of serving a massive, often international audience -- and now Amazon is making its machine learning software open source. The company's Deep Scalable Sparse Tensor Network Engine -- otherwise known as DSSTNE and pronounced "destiny" -- is now available to anyone who's interested in tinkering with it. Amazon hopes that outside influences will help make the platform even more powerful than it already is, according to a report from Engadget. "DSSTNE is built for production deployment of real-world deep learning applications, emphasizing speed and scale over experimental flexibility," reads documentation that accompanies the files released by Amazon. Internally, DSSTNE is used to deliver purchase recommendations to consumers based on their order histories. Product recommendations are big business for Amazon, as having such a daunting catalog of merchandise is really rather worthless unless customers are able to discover items that are relevant to their interests.