Despite its struggles in some countries, virtual reality has steadily grown in popularity throughout China over the last few years, so Microsoft -- better known for its augmented reality headset HoloLens than its VR initiatives -- is getting in on the action. As reported by China's Xinhua news agency, the company today launched the Nanchang City AI VR Innovation Center, a cloud and mobile technology incubator geared towards startups and established manufacturing companies. Though the center's name suggests that Microsoft will focus equally on artificial intelligence and virtual reality initiatives, the incubator was apparently designed to cater to the VR industry in the Jiangxi province, which created a first-of-kind industrial base within China for VR technologies. The local government expects that Microsoft's incubator will "lure dozens of AI, VR and other tech companies" into Nanchang City's Honggutan New District, and expects it to be used to train and support local companies. Microsoft's interest in AI has been strong and clear over the past year, most recently spanning everything from AI-powered suggestions in Microsoft 365 to AI-focused ecological research grants and general availability of Azure Machine Learning.

Simulation software that can create accurate "digital twins" of entire cities is enabling planners, designers and engineers to improve their designs and measure the effect changes will have on the lives of citizens. Cities are hugely complex and dynamic creations. Think about all the parts: millions of people, schools, offices, shops, parks, utilities, hospitals, homes and transport systems. Changing one aspect affects many others. Which is why planning is such a hard job.

AI: Saying no to fear and yes to functionality We are entering into an era where automation will be expected. Employees will expect to have relevant news at the top of their news feed, they will expect to be able to search with a few keywords and for their computer to anticipate the results they need. At the roots of this expectation are the advances in Artificial intelligence that are making it possible. Since the term was first coined in 1956, artificial intelligence has been the subject of fear, fantasy and extensive research. What began in the imagination of Sci-Fi novelists and filmmakers is rapidly becoming reality.

The proliferation of fake news on social media has opened up new directions of research for timely identification and containment of fake news, and mitigation of its widespread impact on public opinion. While much of the earlier research was focused on identification of fake news based on its contents or by exploiting users' engagements with the news on social media, there has been a rising interest in proactive intervention strategies to counter the spread of misinformation and its impact on society. In this survey, we describe the modern-day problem of fake news and, in particular, highlight the technical challenges associated with it. We discuss existing methods and techniques applicable to both identification and mitigation, with a focus on the significant advances in each method and their advantages and limitations. In addition, research has often been limited by the quality of existing datasets and their specific application contexts. To alleviate this problem, we comprehensively compile and summarize characteristic features of available datasets. Furthermore, we outline new directions of research to facilitate future development of effective and interdisciplinary solutions.

Playlist recommendation involves producing a set of songs that a user might enjoy. We investigate this problem in three cold-start scenarios: (i) cold playlists, where we recommend songs to form new personalised playlists for an existing user; (ii) cold users, where we recommend songs to form new playlists for a new user; and (iii) cold songs, where we recommend newly released songs to extend users' existing playlists. We propose a flexible multitask learning method to deal with all three settings. The method learns from user-curated playlists, and encourages songs in a playlist to be ranked higher than those that are not by minimising a bipartite ranking loss. Inspired by an equivalence between bipartite ranking and binary classification, we show how one can efficiently approximate an optimal solution of the multitask learning objective by minimising a classification loss. Empirical results on two real playlist datasets show the proposed approach has good performance for cold-start playlist recommendation.

Artificial intelligence is an umbrella term for various data analytics processes. Machine learning is a subset of AI, and refers to a process that uses statistical techniques to give computers the ability to "learn" with data, without explicitly programming the computer to do so. And, deep learning is a subset, in turn, of machine learning, denoting processes based on deciphering the significance and meaning to be derived, if any, from the input data. According to IHS Markit, there are just two applications where artificial intelligence is used currently in home-security systems. The first is in systems integrated with consumer video cameras.

The style transfer craze kicked off by an app called Prisma a couple of years ago led to a tsunami of painterly selfies flooding social feeds for several months, as we reported at the time, before the rapacious, face-snapping hoards shifted their attention toward fresh spectacles. The same tech is now creeping into (paper) kids' books, via a partnership between children's publisher startup, Kabook, and Prisma Labs: aka the b2b entity that the original app makers pivoted to in late 2017. So instead of AI sending robots into a human-slaying frenzy, per the usual dystopian sci-fi storyline, we find ourselves confronted with neural nets being used to serve up contextual illustrations of children so parents can gift personalized books that seamlessly insert a child's likeness into the story, thereby casting them as a character in the tale. Well, not unless you view this kind of self-centered content manipulation as a threat to children's imaginations and developing sense of empathy. The Kabook integration is the first consumer product partnership that Prisma Labs has scored, according to a press release from the pair.

That means the pooling layer computes a feature vector of size 128 which is passed into dense layers of the feedforward network as we mentioned above. The overall structure of the DNN can be understood as a preprocessor defined in the first part that is being trained to map text sequences into feature vectors in such a way that the weights of the second part can be trained to obtain optimal classification results from the overall network. More details on the implementation and text preprocessing can be found in my GitHub repository for this project. I trained this network for 10 epochs with a batch size of 128 using an 80-20 training/hold-out set. A couple of notes on additional parameters: The vast majority of documents in this collection is of length 5000 or less. So for the maximum input sequence length for the DNN I chose 5000 words. There are roughly 100,000 unique words in this collection of documents. I arbitrarily limited the dictionary that the DNN can learn to 25% of that: 25,000 words. Finally, for the embedding dimension, I chose 300 simply because that is the default embedding dimension for both word2vec and GloVe.

As artificial intelligence moves into the courtroom, much has been written about sentencing algorithms with hidden biases.