In this paper, we study how to make clustering benefiting from differentiable programming whose basic idea is treating the neural network as a language instead of a machine learning method. To this end, we recast the vanilla $k$-means as a novel feedforward neural network in an elegant way. Our contribution is two-fold. On the one hand, the proposed \textit{k}-meansNet is a neural network implementation of the vanilla \textit{k}-means, which enjoys four advantages highly desired, i.e., robustness to initialization, fast inference speed, the capability of handling new coming data, and provable convergence. On the other hand, this work may provide novel insights into differentiable programming. More specifically, most existing differentiable programming works unroll an \textbf{optimizer} as a \textbf{recurrent neural network}, namely, the neural network is employed to solve an existing optimization problem. In contrast, we reformulate the \textbf{objective function} of \textit{k}-means as a \textbf{feedforward neural network}, namely, we employ the neural network to describe a problem. In such a way, we advance the boundary of differentiable programming by treating the neural network as from an alternative optimization approach to the problem formulation. Extensive experimental studies show that our method achieves promising performance comparing with 12 clustering methods on some challenging datasets.

Generative Adversarial Networks (GANs) have shown great capacity on image generation, in which a discriminative model guides the training of a generative model to construct images that resemble real images. Recently, GANs have been extended from generating images to generating sequences (e.g., poems, music and codes). Existing GANs on sequence generation mainly focus on general sequences, which are grammar-free. In many real-world applications, however, we need to generate sequences in a formal language with the constraint of its corresponding grammar. For example, to test the performance of a database, one may want to generate a collection of SQL queries, which are not only similar to the queries of real users, but also follow the SQL syntax of the target database. Generating such sequences is highly challenging because both the generator and discriminator of GANs need to consider the structure of the sequences and the given grammar in the formal language. To address these issues, we study the problem of syntax-aware sequence generation with GANs, in which a collection of real sequences and a set of pre-defined grammatical rules are given to both discriminator and generator. We propose a novel GAN framework, namely TreeGAN, to incorporate a given Context-Free Grammar (CFG) into the sequence generation process. In TreeGAN, the generator employs a recurrent neural network (RNN) to construct a parse tree. Each generated parse tree can then be translated to a valid sequence of the given grammar. The discriminator uses a tree-structured RNN to distinguish the generated trees from real trees. We show that TreeGAN can generate sequences for any CFG and its generation fully conforms with the given syntax. Experiments on synthetic and real data sets demonstrated that TreeGAN significantly improves the quality of the sequence generation in context-free languages.

Is this article written by a bot? In a few years, it might be hard to say one way or another. Artificial intelligence is being applied to many different industries, and the areas of news and journalism are certainly no exception. In fact, automated journalism is already helping create news articles and enhance storytelling. The Washington Post reported last year that its own AI bot, known as Heliograf, published 850 stories entirely autonomously, primarily reporting on sports and the outcomes of regional political races.

The ability to witness non-local correlations lies at the core of foundational aspects of quantum mechanics and its application in the processing of information. Commonly, this is achieved via the violation of Bell inequalities. Unfortunately, however, their systematic derivation quickly becomes unfeasible as the scenario of interest grows in complexity. To cope with that, we propose here a machine learning approach for the detection and quantification of non-locality. It consists of an ensemble of multilayer perceptrons blended with genetic algorithms achieving a high performance in a number of relevant Bell scenarios. Our results offer a novel method and a proof-of-principle for the relevance of machine learning for understanding non-locality.

The paper presents a knowledge representation language $\mathcal{A}log$ which extends ASP with aggregates. The goal is to have a language based on simple syntax and clear intuitive and mathematical semantics. We give some properties of $\mathcal{A}log$, an algorithm for computing its answer sets, and comparison with other approaches.

Microsoft and Amazon will bring Alexa and Cortana to all Echo speakers and Windows 10 users in the U.S. today. As part of a partnership between the Seattle-area tech giants, you can say "Hey Cortana, open Alexa" to Windows 10 PCs and "Alexa, open Cortana" to a range of Echo smart speakers. The public preview bringing the most popular AI assistant on PCs together with the smart speaker with the largest U.S. market share will be available to most people today but will be rolled out to all users in the country over the course of the next week, a Microsoft spokesperson told VentureBeat in an email. Each of the assistants brings unique features to the table. Cortana, for example, can schedule a meeting with Outlook, create location-based reminders, or draw on LinkedIn to tell you about people in your next meeting.

BOGOTÁ, Colombia--With drug crops booming, Colombia's police are busily testing whether drones carrying defoliants can efficiently kill the leaf used to make cocaine and win the support of Trump administration officials concerned about this country's growing capacity to supply drugs to American consumers. Antidrug officials here say that in recent weeks they have deployed 10 drones, each weighing 50 pounds when loaded with herbicide, in southwest Nariño province. The small, remotely guided aircraft destroyed hundreds of acres of coca in a first round of tests, said police and the company contracted by the government to supply the drones. Colombia's new president, Iván Duque, said that he wants some kind of aerial fumigation of coca fields, which expanded 160% to 516,000 acres from 2012 to 2017, the White House reported in June. But he prefers drones over planes to drop the herbicide, which would mitigate damage to legal crops growing adjacent to coca fields.

Sessions are subject to change. Day & Time: Wednesday, 11:00 AM – 11:45 AM, Grand Oaks A&B Presenter: Jim Breen, Doug Foster Need to get your team trained on InterSystems products quickly? Attend this session to learn how you can get your employees up to speed and add value to your company – fast! Hear how other InterSystems' clients have created successful teams using Learning Services content as one piece of the puzzle, and how you can too! Takeaway: InterSystems Learning Services can help me quickly onboard new employees and grow the skill sets of existing employees. Day & Time: Monday, 2:00 PM – 2:45 PM, Grand Oaks E&F Tuesday, 2:00 PM – 2:45 PM, Grand Oaks C&D Presenter: Andreas Dieckow This session provides an overview of what it takes to move an existing Caché or Ensemble application to InterSystems IRIS Data Platform. You will learn that migration is not urgent (unless you want to take advantage of new features in InterSystems IRIS) but that it is often less complex than you might expect.

I sent Dr. Reina Reyes a Mashable article which posited that "Mission: Impossible" movies are more successful the more Tom Cruise runs. It claimed positive correlation between quality (Rotten Tomatoes review scores) and distance run (estimated at 14.6 feet per second of screen time). Before "M:I-Fallout," Cruise ran over 3,000 feet in both top-rated M:I movies, "M:I-III" and "Ghost Protocol." Reyes quickly distilled it into a scatter plot on Facebook, complete with a still of Cruise running. I am blessed with friends who continually teach me data visualization and big data's other languages.

Nefarious actors on social media and other platforms often spread rumors and falsehoods through images whose metadata (e.g., captions) have been modified to provide visual substantiation of the rumor/falsehood. This type of modification is referred to as image repurposing, in which often an unmanipulated image is published along with incorrect or manipulated metadata to serve the actor's ulterior motives. We present the Multimodal Entity Image Repurposing (MEIR) dataset, a substantially challenging dataset over that which has been previously available to support research into image repurposing detection. The new dataset includes location, person, and organization manipulations on real-world data sourced from Flickr. We also present a novel, end-to-end, deep multimodal learning model for assessing the integrity of an image by combining information extracted from the image with related information from a knowledge base. The proposed method is compared against state-of-the-art techniques on existing datasets as well as MEIR, where it outperforms existing methods across the board, with AUC improvement up to 0.23.