Wix.com Inc, the creator of a cloud-based web development platform, has announced that it will be offering a new web design tool called Wix ADI (Artificial Design Intelligence), which the company says is capable of building websites tailored to individual users. The new tool works by pulling in curated layouts, images, text, forms, and other website features that Wix says will be customized to the needs of each user. In a release, Wix said that its AI has built its algorithms based on 86 million users, and the company said the Wix ADI will "ensure every website design will be stunning and complete." "Wix ADI is the world's first technology platform that combines website design and content creation with artificial intelligence to enable complete websites to be created in a matter of minutes," said Avishai Abrahami, Co-Founder and CEO of Wix. "Building a professional and complete website is the first step in getting online, and while Wix today makes it easy, challenges still exist. With Wix ADI, we solve these challenges, getting businesses online with a fully functioning, stunning website easier and faster than ever. Technology innovation is Wix's core competency, and for nearly a decade, small businesses have relied on Wix not only to build websites but also to manage and grow their businesses online."

We've been using TensorFlow in daily research and engineering since it was released almost six months ago. We've learned a lot of things along the way. Because there are many subjective articles on TensorFlow and not enough helpful documentation, I've sprinkled in examples, tutorials, docs, and code snippets wherever possible. When it comes to machine learning, it is easy to focus on the tech (features, capabilities, benchmarks, etc). But good programmers know it is much harder to write code that humans will use, versus code that a machine can compile and execute.

In a blog post titled "Preparing for the Future of Artificial Intelligence," Deputy U.S. CTO Ed Felten announced the creation of the Machine Learning and Artificial Intelligence Subcommittee and a series of workshops to encourage public discourse about the emerging technology. "There are tremendous opportunities and an array of considerations across the Federal Government in privacy, security, regulation, law, and research and development to be taken into account when effectively integrating this technology into both government and private-sector activities," wrote Felten. Related: Google's newly launched Magenta Project aims to create art with artificial intelligence Open to the public, the workshops are co-hosted by the White House Office of Science and Technology and National Economic Council with academic and non-profit organizations across the country. The goal is to "identify challenges and opportunities" related to AI, which will lead to a public report by the subcommittee later this year. The first event was co-hosted by the University of Washington's School of Law in Seattle on May 24 and considered AI as applied to law and governance.

Dynamic functional connectivity (FC) has in recent years become a topic of interest in the neuroimaging community. Several models and methods exist for both functional magnetic resonance imaging (fMRI) and electroencephalography (EEG), and the results point towards the conclusion that FC exhibits dynamic changes. The existing approaches modeling dynamic connectivity have primarily been based on time-windowing the data and k-means clustering. We propose a non-parametric generative model for dynamic FC in fMRI that does not rely on specifying window lengths and number of dynamic states. Rooted in Bayesian statistical modeling we use the predictive likelihood to investigate if the model can discriminate between a motor task and rest both within and across subjects. We further investigate what drives dynamic states using the model on the entire data collated across subjects and task/rest. We find that the number of states extracted are driven by subject variability and preprocessing differences while the individual states are almost purely defined by either task or rest. This questions how we in general interpret dynamic FC and points to the need for more research on what drives dynamic FC.

The counting grid is a grid of microtopics, sparse word/feature distributions. The generative model associated with the grid does not use these microtopics individually, but in predefined groups which can only be (ad)mixed as such. Each allowed group corresponds to one of all possible overlapping rectangular windows into the grid. The capacity of the model is controlled by the ratio of the grid size and the window size. This paper builds upon the basic counting grid model and it shows that hierarchical reasoning helps avoid bad local minima, produces better classification accuracy and, most interestingly, allows for extraction of large numbers of coherent microtopics even from small datasets. We evaluate this in terms of consistency, diversity and clarity of the indexed content, as well as in a user study on word intrusion tasks. We demonstrate that these models work well as a technique for embedding raw images and discuss interesting parallels between hierarchical CG models and other deep architectures.

We consider online optimization in the 1-lookahead setting, where the objective does not decompose additively over the rounds of the online game. The resulting formulation enables us to deal with non-stationary and/or long-term constraints , which arise, for example, in online display advertising problems. We propose an on-line primal-dual algorithm for which we obtain dynamic cumulative regret guarantees. They depend on the convexity and the smoothness of the non-additive penalty, as well as terms capturing the smoothness with which the residuals of the non-stationary and long-term constraints vary over the rounds. We conduct experiments on synthetic data to illustrate the benefits of the non-additive penalty and show vanishing regret convergence on live traffic data collected by a display advertising platform in production.

A good clustering can help a data analyst to explore and understand a data set, but what constitutes a good clustering may depend on domain-specific and application-specific criteria. These criteria can be difficult to formalize, even when it is easy for an analyst to know a good clustering when she sees one. We present a new approach to interactive clustering for data exploration, called \ciif, based on a particularly simple feedback mechanism, in which an analyst can choose to reject individual clusters and request new ones. The new clusters should be different from previously rejected clusters while still fitting the data well. We formalize this interaction in a novel Bayesian prior elicitation framework. In each iteration, the prior is adapted to account for all the previous feedback, and a new clustering is then produced from the posterior distribution. To achieve the computational efficiency necessary for an interactive setting, we propose an incremental optimization method over data minibatches using Lagrangian relaxation. Experiments demonstrate that \ciif can produce accurate and diverse clusterings.

Robots are becoming more commonplace in human environments, such as schools, homes, hospitals, and work settings, and are expected to accomplish a wide variety of tasks. Given the near infinite number of tasks robots might be expected to perform in these varied settings, it is not feasible for robot designers to completely pre-program machines before they are deployed. Many researchers have suggested this problem can be addressed via end-user robot programming, where users can modify and create new behaviors for their robot to best suit their needs and preferences [2], [1]. Learning from demonstration (LfD) is one such method that enables people to readily develop custom robot behavior [2]. In LfD, a learner automatically creates a mapping between states and actions by watching a teacher perform the task; the learner can then replicate the teacher's actions. The main benefit of LfD is that it is an intuitive way for people to teach robots and does not require the teacher to have highly specialized knowledge, such as the ability to directly program the robot [3]. There has been significant research in how to design and implement LfD systems, including how people want to teach robots. Work by Thomaz et al. [23] showed that LfD systems could be improved for both the teacher and learner if greater communicative channels could be employed during the learning process. We build upon this work, and specifically are interested in ways to enable human teachers to have more efficient and naturalistic interactions, by way of a common human-human interaction (HHI) phenomena: grounding sequences.

In decentralized networks (of sensors, connected objects, etc.), there is an important need for efficient algorithms to optimize a global cost function, for instance to learn a global model from the local data collected by each computing unit. In this paper, we address the problem of decentralized minimization of pairwise functions of the data points, where these points are distributed over the nodes of a graph defining the communication topology of the network. This general problem finds applications in ranking, distance metric learning and graph inference, among others. We propose new gossip algorithms based on dual averaging which aims at solving such problems both in synchronous and asynchronous settings. The proposed framework is flexible enough to deal with constrained and regularized variants of the optimization problem. Our theoretical analysis reveals that the proposed algorithms preserve the convergence rate of centralized dual averaging up to an additive bias term. We present numerical simulations on Area Under the ROC Curve (AUC) maximization and metric learning problems which illustrate the practical interest of our approach.

Learning robust value functions given raw observations and rewards is now possible with model-free and model-based deep reinforcement learning algorithms. There is a third alternative, called Successor Representations (SR), which decomposes the value function into two components -- a reward predictor and a successor map. The successor map represents the expected future state occupancy from any given state and the reward predictor maps states to scalar rewards. The value function of a state can be computed as the inner product between the successor map and the reward weights. In this paper, we present DSR, which generalizes SR within an end-to-end deep reinforcement learning framework. DSR has several appealing properties including: increased sensitivity to distal reward changes due to factorization of reward and world dynamics, and the ability to extract bottleneck states (subgoals) given successor maps trained under a random policy. We show the efficacy of our approach on two diverse environments given raw pixel observations -- simple grid-world domains (MazeBase) and the Doom game engine.