The "willing suspension of disbelief" is the idea that we (the audience, readers, viewers, content consumers) are willing to suspend judgment about the implausibility of the narrative for the quality of our own enjoyment. We do it all the time. Two-dimensional video on our screens is smaller than life and flat and not in real time, but we ignore those facts and immerse ourselves in the stories as if they were real. We have also learned the "conventions" of each medium. While we watch a movie or a video, we don't yell to the characters on the screen "Duck!" or "Look out!" when something is about to happen to them.

Thanks to the wonders of AI, it's possible to ask a robot to help you figure out the best way to deal with a difficult situation. One of the greatest hits as far as Christmas gift-giving was a gizmo called an Echo Dot I gave my nephew, Nathan. It was a hit despite the obligatory annoying learning curve that comes when any Southerner tries to talk to an artificial intelligence device ("Alexer, G-darnit, stop giving me the temperature in Celsius, I want it in American!") I had arrived in Chapel Hill weary from a verbal battle with Siri who, based on the convoluted traffic pattern she recommended, is a huge Duke fan. Artificial intelligence is a big buzzword for 2017.

Next time you stop for gas at a self-serve pump, say hello to the robot in front of you. Its life story can tell you a lot about the robot economy roaring toward us like an EF5 tornado on the prairie. Yeah, your automated gas pump killed a lot of jobs over the years, but its biography might give you hope that the coming wave of automation driven by artificial intelligence (AI) will turn out better for almost all of us than a lot of people seem to think. The first crude version of an automated gas-delivering robot appeared in 1964 at a station in Westminster, Colorado. Short Stop convenience store owner John Roscoe bought an electric box that let a clerk inside activate any of the pumps outside. Self-serve pumps didn't catch on until the 1970s, when pump-makers added automation that let customers pay at the pump, and over the next 30 years, stations across the nation installed these task-specific robots and fired attendants. By the 2000s, the gas attendant job had all but disappeared.

Amazon's 2016 has been record breaking on many fronts. The company recorded its sixth consecutive quarterly profit (previously, it mostly hemorrhaged cash). Meanwhile, this year marked Amazon's growing strength in hardware with its hit Echo home automation hub Amazon Echo, and its companion voice assistant Alexa. The company has also become force in entertainment, debuting a line of hit original shows through its Amazon Video Prime service. It's hard to imagine how Amazon could top 2016, but here are some likely moves by the Seattle-based Goliath in 2017: To save money over the past year, Amazon has been seeking to take over more shipping duties from the likes of UPS and FedEx by leasing trucks, planes, and ships.

This paper deals with price optimization, which is to find the best pricing strategy that maximizes revenue or profit, on the basis of demand forecasting models. Though recent advances in regression technologies have made it possible to reveal price-demand relationship of a number of multiple products, most existing price optimization methods, such as mixed integer programming formulation, cannot handle tens or hundreds of products because of their high computational costs. To cope with this problem, this paper proposes a novel approach based on network flow algorithms. We reveal a connection between supermodularity of the revenue and cross elasticity of demand. On the basis of this connection, we propose an efficient algorithm that employs network flow algorithms. The proposed algorithm can handle hundreds or thousands of products, and returns an exact optimal solution under an assumption regarding cross elasticity of demand. Even in case in which the assumption does not hold, the proposed algorithm can efficiently find approximate solutions as good as can other state-of-the-art methods, as empirical results show.

We consider the assortment optimization problem when customer preferences follow a mixture of Mallows distributions. The assortment optimization problem focuses on determining the revenue/profit maximizing subset of products from a large universe of products; it is an important decision that is commonly faced by retailers in determining what to offer their customers. There are two key challenges: (a) the Mallows distribution lacks a closed-form expression (and requires summing an exponential number of terms) to compute the choice probability and, hence, the expected revenue/profit per customer; and (b) finding the best subset may require an exhaustive search. Our key contributions are an efficiently computable closed-form expression for the choice probability under the Mallows model and a compact mixed integer linear program (MIP) formulation for the assortment problem.

We propose Sketched Online Newton (SON), an online second order learning algorithm that enjoys substantially improved regret guarantees for ill-conditioned data. SON is an enhanced version of the Online Newton Step, which, via sketching techniques enjoys a running time linear in the dimension and sketch size. We further develop sparse forms of the sketching methods (such as Oja's rule), making the computation linear in the sparsity of features. Together, the algorithm eliminates all computational obstacles in previous second order online learning approaches.

Time series prediction problems are becoming increasingly high-dimensional in modern applications, such as climatology and demand forecasting. For example, in the latter problem, the number of items for which demand needs to be forecast might be as large as 50,000. In addition, the data is generally noisy and full of missing values. Thus, modern applications require methods that are highly scalable, and can deal with noisy data in terms of corruptions or missing values. However, classical time series methods usually fall short of handling these issues. In this paper, we present a temporal regularized matrix factorization (TRMF) framework which supports data-driven temporal learning and forecasting. We develop novel regularization schemes and use scalable matrix factorization methods that are eminently suited for high-dimensional time series data that has many missing values. Our proposed TRMF is highly general, and subsumes many existing approaches for time series analysis. We make interesting connections to graph regularization methods in the context of learning the dependencies in an autoregressive framework. Experimental results show the superiority of TRMF in terms of scalability and prediction quality. In particular, TRMF is two orders of magnitude faster than other methods on a problem of dimension 50,000, and generates better forecasts on real-world datasets such as Wal-mart E-commerce datasets.

Innovation that lets firms reach out and touch their customers is moving at warp speed, building on expectations that were unthinkable just five years ago. When today's consumer has questions about your product, it is no longer acceptable to wait for the answers -- they must be addressed in real time or the customer will turn to the next readily available and better option. How can your organization convert this impatient prospect into a customer and lay the groundwork for loyalty as we know it today? Our baker's dozen of customer experience trends for 2017 suggest a number of great starting points. The survey-driven Customer Satisfaction Score (CSAT), Net Promoter Score (NPS) and Customer Effort Score (CES) each get at different pieces of the puzzle, but all of them oversimplify.

Alexa and Siri become new retail customers: Home-based digital assistants like Amazon's Alexa and Apple's Siri will proliferate and expand into eCommerce, enabling consumers to'shop by voice'. These assistants are becoming pervasive as the ecosystems expand outside the home; Amazon is already integrating Alexa into Ford, BMW, and Hyundai autos and Apple is a growing force in cars. IoT hits a tipping point: The Internet of Things (IoT) has been talked about a lot for both in-home and in-store applications. Omnipresent omni-channel: Platforms are being put in place that provide consistent relevant content across every digital touchpoint while ensuring a cohesive and comprehensive user experience across devices and channels both inside and outside the store. Relevancy required: Strategic hyper-personalization goes from a'nice to have' to a necessity to keep customers (especially millennials and younger shoppers) engaged and to make the most efficient use of marketing budgets.