WGAN has really paved the way for alot of GAN applications both in images and text. However, one problem I primarily see with training WGAN for text is that the generator fails to fully converge. That is, the wasserstein distance still remains large and despite numerous steps, the generator will not converge any further. To aid the generator, one idea is to allow the generator to see the Discriminator's preactivations from hidden layers and allow the generator to revise its outputs. The idea here is that the generator gets a chance to propose a sequence, see how the discriminator will evaluate it, and revise it sequence all in one differentiable calculation.

Apple's most expensive smartphone, the £999 iPhone X, is almost ready to land in stores and a few publications specially selected by Apple have been given early access to the phone. So what do they think? Is the iPhone X really the "future of smartphones"? The iPhone X has an all-screen front design with a 5.8in OLED screen, no home button and an odd looking notch at the top for front-facing camera and sensors. The rest of the device resembles the the iPhone 8 and 8 Plus, with a dual camera on the glass back.

Sometimes the scariest place to be is your own mind. Shelley is an AI program that generates the beginnings of horror stories, and it's trained by original horror fiction posted to Reddit. Designed by researchers from MIT Media Lab, Shelley launched on Twitter on Oct. 21. The team behind Shelley is hoping to learn more about how machines can evoke emotional responses in humans. "The rapid progress in the field of Artificial Intelligence (AI) has people worried about everything from mass unemployment to the annihilation of the human race at the hand of evil robots," writes researcher Iyad Rahwan by email.

Alex Karpenko hands me a camera and tells me to run. We're standing on a pier in San Francisco, and the device in Karpenko's hand is an unreleased prototype of a new, software-driven video camera called Rylo. Karpenko wants me to see what he and co-founder Chris Cunningham show recruits and investors when they ask why they should get involved. Karpenko says I don't have to worry about where to point the camera, or try to hold it still. So I grab the camera--a small, oblong 360-degree shooter with a lens on either side--and start running.

Should musicians, whose job relies on the unpredictable and mysterious workings of human imagination, heed the warnings about artificial intelligence driving humans into unemployment? Apparently, the answer to that question is bit complicated. Artificial intelligence is gradually transforming into a general-purpose technology that permeates virtually every aspect of human life and society. Just like electricity (another general-purpose technology) which had an impact on music and musical instruments, AI algorithms will inevitably change the way we create and perform music. While this does not mean an end to the era of human musicians, it is fair to say that some dramatic changes are lying ahead.

Should musicians, whose job relies on the unpredictable and mysterious workings of human imagination, heed the warnings about artificial intelligence driving humans into unemployment? Apparently, the answer to that question is bit complicated. Artificial intelligence is gradually transforming into a general-purpose technology that permeates virtually every aspect of human life and society. Just like electricity (another general-purpose technology) which had an impact on music and musical instruments, AI algorithms will inevitably change the way we create and perform music. While this does not mean an end to the era of human musicians, it is fair to say that some dramatic changes are lying ahead. At the heart of recent artificial intelligence breakthroughs are machine learning algorithms, programs that find patterns in large sets of data.

It's the most anticipated smartphone in years, and its success will be important for Apple. I was one of just a few journalists to be given a chance to spend the last nine days getting to grips with the new iPhone X (written X but pronounced 10). There's quite a bit to say. This is one of the key features on the new phone and it's a winner. I've gone into more detail separately here but the essence is that the Home Button found on all previous iPhones, and the fingerprint sensor, Touch ID, have been removed. Now, a sophisticated facial recognition system means that when you look at the phone, if it recognises you, it unlocks. Other phone manufacturers have included facial recognition but this one is significantly more consistent and reliable.

My new favorite vlog to follow is that of Simone Giertz. She is a Swedish inventor, maker and a robotics enthusiast. She is your real life, modern and feminine version of Dr. Emmet Brown (from the movie Back to the Future). Even better, she builds artificial intelligence/robots to help her with the mundane daily tasks like brushing teeth or making a bowl of cereal. I doubt that these robots are truly useful, but they are funny.

Most people do not interact with Semantic Web data directly. Unless they have the expertise to understand the underlying technology, they need textual or visual interfaces to help them make sense of it. We explore the problem of generating natural language summaries for Semantic Web data. This is non-trivial, especially in an open-domain context. To address this problem, we explore the use of neural networks. Our system encodes the information from a set of triples into a vector of fixed dimensionality and generates a textual summary by conditioning the output on the encoded vector. We train and evaluate our models on two corpora of loosely aligned Wikipedia snippets and DBpedia and Wikidata triples with promising results.

Research[R] LSTM as a Dynamically Computed Element-wise Weighted Sum - reinterprets gating in LSTMs as self-attention over time (a weighted sum over the candidate states c _t); shows that dependence on h_t-1 for computing c _t is not necessary; thus gating does the heavy lifting in LSTMs, not h h mappings (openreview.net)