In the era of big data, logistic planning can be made data-driven to take advantage of accumulated knowledge in the past. While in the movie industry, movie planning can also exploit the existing online movie knowledge library to achieve better results. However, it is ineffective to solely rely on conventional heuristics for movie planning, due to a large number of existing movies and various real-world factors that contribute to the success of each movie, such as the movie genre, available budget, production team (involving actor, actress, director, and writer), etc. In this paper, we study a "Blockbuster Planning" (BP) problem to learn from previous movies and plan for low budget yet high return new movies in a totally data-driven fashion. After a thorough investigation of an online movie knowledge library, a novel movie planning framework "Blockbuster Planning with Maximized Movie Configuration Acquaintance" (BigMovie) is introduced in this paper. From the investment perspective, BigMovie maximizes the estimated gross of the planned movies with a given budget. It is able to accurately estimate the movie gross with a 0.26 mean absolute percentage error (and 0.16 for budget). Meanwhile, from the production team's perspective, BigMovie is able to formulate an optimized team with people/movie genres that team members are acquainted with. Historical collaboration records are utilized to estimate acquaintance scores of movie configuration factors via an acquaintance tensor. We formulate the BP problem as a non-linear binary programming problem and prove its NP-hardness. To solve it in polynomial time, BigMovie relaxes the hard binary constraints and addresses the BP problem as a cubic programming problem. Extensive experiments conducted on IMDB movie database demonstrate the capability of BigMovie for an effective data-driven blockbuster planning.

The first testing sessions for SampleRNN – an artificially intelligent software developed by computer scientist duo CJ Carr and Zach Zukowski, AKA Dadabots – sounded more like a screamo gig than a machine-learning experiment. Carr and Zukowski hoped their program could generate full-length black metal and math rock albums by feeding it small chunks of sound. The first trial consisted of encoding and entering in a few Nirvana a cappellas. "When it produced its first output," Carr tells me over email, "I was expecting to hear silence or noise because of an error we made, or else some semblance of singing. The first thing it did was scream about Jesus. We looked at each other like, 'What the fuck?'"

If at first you don't succeed, slash your price and try, try again. That's the thinking now being applied by Caavo, the company which launched a new-age universal TV device back in February. I reviewed it and liked many things about it, but I also complained about its $400 price, difficult setup, and many quirks, including a lack of HDR support. Someone over at Caavo HQ was listening. Less than eight months later, that bulky $400 Caavo has vanished.

Google brings video to the talking speaker category with the new Google Home Hub. USA TODAY's Jefferson Graham explains why the device has potential. Google officially released the Home Hub, a $149 digital photo frame, YouTube video player, and visual recipe reciter (among other things) that looks to take on the second edition of the $229 Amazon Echo Show, which has been in release but is sold out through November 5th. The Show looks to bring the popular Alexa personal assistant into the video age, with live TV from Hulu, video-on-command from Amazon Prime Video, visual recipes, digital photos and Alexa smarts. The two are expected to be among the most heavily marketed new holiday products.

Google Home Hub combines a Google Home digital assistant with a 7-inch screen, allowing a graphic display of your schedule, weather, and commute traffic all at the command of your voice. There's more in there too: It has a good speaker for playing music, will act as a digital photo frame when not otherwise in use, and you can use it to watch your favorite shows if you have a compatible pay TV service, or monitor the feed from your home security cameras. In many ways it mirrors what a modern Android phone can already do, but with a better speaker and a bigger, always-on screen. There's a lot to like about this impressive little gadget, and its thoughtfully designed software is a big improvement on a smartphone. The Home Hub will almost certainly surprise you with its small size.

The Home Hub is Google's first own-brand smart display, combining Google Assistant, advanced smart-home control and a digital photo frame into a neat and tidy package. Google isn't the first to market with smart displays. Amazon's Echo Show put the company's Alexa on a screen a year ago, while Google Assistant smart displays made by Harman, Lenovo and LG were released a few months ago. But Google's Home Hub is slightly different. Firstly, it doesn't have a camera on it, which Google goes to great lengths to repeatedly tell you in the hope you will find it less creepy and feel more comfortable putting it in places like your bedroom.

Imagine if we could call out to our living room TV to access a show or movie, get step-by-step cooking directions, watch YouTube clips on demand, have it double as a digital photo frame, look up anything in the world, make phone calls and run your home automation system. How cool would that be? That's the promise of Google's new Home Hub, a $149 talking speaker that is taking on Amazon's $229 Echo Show in what's expected to be one of the biggest marketing battles of the holidays. How small is the Google Home Hub? So tiny that the Amazon Echo dwarfs it, and Mr. Jinx the cat towers over it.

I have heard that both feed-forward networks as well as recurrent neural networks are universal function approximators, i.e., they can approximate any function arbitrarily close with increasing hidden layers and hidden units. Is there a difference between smooth and non-smooth functions? What is better, a few hidden layers with many hidden units per layer or deep learning, i.e., many hidden layers? Has anyone tried to use LSTM networks to do this kind of stuff?

Articles that might be of interest to influencers. Inc.com--Google's spent the last 10 years studying the habits of effective managers. You can learn a lot from their conclusions. October 21, 2018 Go to Inc.com Small Biz Daily--To people outside the digital world, the term "hack" implies either a security breach or sleight of hand. You can't do as much as you think you can.