Classification and regression in which the inputs are graphs of arbitrary size and shape have been paid attention in various fields such as computational chemistry and bioinformatics. Subgraph indicators are often used as the most fundamental features, but the number of possible subgraph patterns are intractably large due to the combinatorial explosion. We propose a novel efficient algorithm to jointly learn relevant subgraph patterns and nonlinear models of their indicators. Previous methods for such joint learning of subgraph features and models are based on search for single best subgraph features with specific pruning and boosting procedures of adding their indicators one by one, which result in linear models of subgraph indicators. In contrast, the proposed approach is based on directly learning regression trees for graph inputs using a newly derived bound of the total sum of squares for data partitions by a given subgraph feature, and thus can learn nonlinear models through standard gradient boosting. An illustrative example we call the Graph-XOR problem to consider nonlinearity, numerical experiments with real datasets, and scalability comparisons to naive approaches using explicit pattern enumeration are also presented.

Over the last decade, digital media (web or app publishers) generalized the use of real time ad auctions to sell their ad spaces. Multiple auction platforms, also called Supply-Side Platforms (SSP), were created. Because of this multiplicity, publishers started to create competition between SSPs. In this setting, there are two successive auctions: a second price auction in each SSP and a secondary, first price auction, called header bidding auction, between SSPs.In this paper, we consider an SSP competing with other SSPs for ad spaces. The SSP acts as an intermediary between an advertiser wanting to buy ad spaces and a web publisher wanting to sell its ad spaces, and needs to define a bidding strategy to be able to deliver to the advertisers as many ads as possible while spending as little as possible. The revenue optimization of this SSP can be written as a contextual bandit problem, where the context consists of the information available about the ad opportunity, such as properties of the internet user or of the ad placement.Using classical multi-armed bandit strategies (such as the original versions of UCB and EXP3) is inefficient in this setting and yields a low convergence speed, as the arms are very correlated. In this paper we design and experiment a version of the Thompson Sampling algorithm that easily takes this correlation into account. We combine this bayesian algorithm with a particle filter, which permits to handle non-stationarity by sequentially estimating the distribution of the highest bid to beat in order to win an auction. We apply this methodology on two real auction datasets, and show that it significantly outperforms more classical approaches.The strategy defined in this paper is being developed to be deployed on thousands of publishers worldwide.

For sufficiently large datasets, neural engines tend to outshine statistical models in most natural language processing problems. However, a speech accent remains a challenge for both approaches. Phonologists manually create general rules describing a speaker's accent, but their results remain underutilized. In this paper, we propose a model that automatically retrieves phonological generalizations from a small dataset. This method leverages the difference in pronunciation between a particular dialect and General American English (GAE) and creates new accented samples of words. The proposed model is able to learn all generalizations that previously were manually obtained by phonologists. We use this statistical method to generate a million phonological variations of words from the CMU Pronouncing Dictionary and train a sequence-to-sequence RNN to recognize accented words with 59% accuracy.

We propose a method to decrease the number of hidden units of the restricted Boltzmann machine while avoiding decrease of the performance measured by the Kullback-Leibler divergence. Then, we demonstrate our algorithm by using numerical simulations.

We consider the problem of inference in discrete probabilistic models, that is, distributions over subsets of a finite ground set. These encompass a range of well-known models in machine learning, such as determinantal point processes and Ising models. Locally-moving Markov chain Monte Carlo algorithms, such as the Gibbs sampler, are commonly used for inference in such models, but their convergence is, at times, prohibitively slow. This is often caused by state-space bottlenecks that greatly hinder the movement of such samplers. We propose a novel sampling strategy that uses a specific mixture of product distributions to propose global moves and, thus, accelerate convergence. Furthermore, we show how to construct such a mixture using semigradient information. We illustrate the effectiveness of combining our sampler with existing ones, both theoretically on an example model, as well as practically on three models learned from real-world data sets.

This paper studies a stylized, yet natural, learning-to-rank problem and points out the critical incorrectness of a widely used nearest neighbor algorithm. We consider a model with $n$ agents (users) $\{x_i\}_{i \in [n]}$ and $m$ alternatives (items) $\{y_j\}_{j \in [m]}$, each of which is associated with a latent feature vector. Agents rank items nondeterministically according to the Plackett-Luce model, where the higher the utility of an item to the agent, the more likely this item will be ranked high by the agent. Our goal is to find neighbors of an arbitrary agent or alternative in the latent space. We first show that the Kendall-tau distance based kNN produces incorrect results in our model. Next, we fix the problem by introducing a new algorithm with features constructed from "global information" of the data matrix. Our approach is in sharp contrast to most existing feature engineering methods. Finally, we design another new algorithm identifying similar alternatives. The construction of alternative features can be done using "local information," highlighting the algorithmic difference between finding similar agents and similar alternatives.

Tornadoes are the most violent of all atmospheric storms. In a typical year, the United States experiences hundreds of tornadoes with associated damages on the order of one billion dollars. Community preparation and resilience would benefit from accurate predictions of these economic losses, particularly as populations in tornado-prone areas continue to increase in density and extent. Here, we use artificial neural networks to predict tornado-induced property damage using publicly available data. We find that the large number of tornadoes which cause zero property damage (30.6% of the data) poses a challenge for predictive models. We developed a model that predicts whether a tornado will cause property damage to a high degree of accuracy (out of sample accuracy = 0.829 and AUROC = 0.873). Conditional on a tornado causing damage, another model predicts the amount of damage. When combined, these two models yield an expected value for the amount of property damage caused by a tornado event. From the best-performing models (out of sample mean squared error = 0.089 and R2 = 0.473), we provide an interactive, gridded map of monthly expected values for the year 2018. One major weakness is that the model predictive power is optimized with log-transformed, mean-normalized property damages, however this leads to large natural-scale residuals for the most destructive tornadoes. The predictive capacity of this model along with an interactive interface may provide an opportunity for science-informed tornado disaster planning.

In 1942, science fiction author Isaac Asimov introduced the world to his three laws of robotics. An incredibly prescient visionary, Asimov started the world thinking about the potential challenges sentient technology might present the world of humanity. In LinkedIn's Financial Services/Fintech survey of more than 1,000 professionals from the broader FI/Fintech space, it is clear that the threats and opportunities associated with A.I. have never been more present conceptually than they are today. When you look at some of the organizations making big bets on A.I. today, the online lists always include technology majors, but we don't yet see banks investing anywhere near the scale of Microsoft, Google, Apple, Alibaba, Baidu and others. Industrial players like Boeing and Tesla are making big bets on A.I., so it is reasonable to expect that we should see big investments coming through financial services also.

Apple's digital assistant left a leading member of the U.K. government red-faced on Tuesday, July 3, after it unexpectedly piped up during a speech he was giving to lawmakers in the British Parliament. Perhaps a little too keen to offer help, Siri interrupted a statement that defense secretary Gavin Williamson was giving to the House of Commons about the situation in Syria. Evidently keeping his phone in always-listening mode, Apple's digital assistant should really only have responded upon hearing "Hey, Siri." But, with his iPhone in his pocket, it seems the word "Syria" prompted the assistant to spring into action. As Williamson addressed lawmakers, Siri got back to the defense secretary with its findings, with the response picked up by the Commons' microphones: "I found something on the web for Syria, Syrian Democratic Forces supported by coalition … " Defence Secretary Gavin Williamson'heckled' by Siri at the despatch box https://t.co/CQlxXm5KAa

WASHINGTON (The Washington Post) – PGA golfers such as four-time major champion Rory McIlroy embrace the tens of thousands of data points – roughly 32,000 per event – that the tour's ShotLink System has offered since 2001. "I made the decision at the end of last year to really look at my stats," McIlroy said after last week's Travelers Championship. "I think they've become very important, and I think the strokes-gained stats, whether it's tee to green or putting or around the green or whatever, I think that's been one of the biggest changes for good that we've seen in golf, because it really just lets you see how your game stacks up against everyone else." For the first time Thursday at the Quicken Loans National at TPC Potomac at Avenel Farm, three fixed, high-resolution cameras, part of the tour's upgraded ShotLink ball-tracking system, replaced the human-operated laser on every green of every hole, capturing the ball in motion as opposed to only the ball at rest. "It's the next phase of how we get the data without having to have human interaction on everything that happens," said Matt Troka, senior vice president of product and partner management of CDW, a technology partner of the PGA Tour.