Recently, there has been a growing interest in the problem of learning rich implicit models - those from which we can sample, but can not evaluate their density. These models apply some parametric function, such as a deep network, to a base measure, and are learned end-to-end using stochastic optimization. One strategy of devising a loss function is through the statistics of two sample tests - if we can fool a statistical test, the learned distribution should be a good model of the true data. However, not all tests can easily fit into this framework, as they might not be differentiable with respect to the data points, and hence with respect to the parameters of the implicit model. Motivated by this problem, in this paper we show how two such classical tests, the Friedman-Rafsky and k-nearest neighbour tests, can be effectively smoothed using ideas from undirected graphical models - the matrix tree theorem and cardinality potentials. Moreover, as we show experimentally, smoothing can significantly increase the power of the test, which might of of independent interest. Finally, we apply our method to learn implicit models.

Stochastic gradient Markov Chain Monte Carlo (SG-MCMC) has been developed as a flexible family of scalable Bayesian sampling algorithms. However, there has been little theoretical analysis of the impact of minibatch size to the algorithm's convergence rate. In this paper, we prove that under a limited computational budget/time, a larger minibatch size leads to a faster decrease of the mean squared error bound (thus the fastest one corresponds to using full gradients), which motivates the necessity of variance reduction in SG-MCMC. Consequently, by borrowing ideas from stochastic optimization, we propose a practical variance-reduction technique for SG-MCMC, that is efficient in both computation and storage. We develop theory to prove that our algorithm induces a faster convergence rate than standard SG-MCMC. A number of large-scale experiments, ranging from Bayesian learning of logistic regression to deep neural networks, validate the theory and demonstrate the superiority of the proposed variance-reduction SG-MCMC framework.

In essence, people who interact with advanced technology want to be able to trust it appropriately, and then act on that trust. In interpersonal relationships, and otherwise, humans act largely based on trust. For example, a supervisor asks a subordinate to accomplish a task based on several factors that indicate they can trust them to accomplish that task. When consumers make purchases, they do so with trust that the product will perform as promised. Likewise, when using something like an autonomous vehicle, the user must be able to trust it appropriately in order to use it properly. With the rapid advancement of the capabilities of intelligent computing technology to do tasks that were previously assumed to be too complicated for computers, there has been much recent discussion regarding how humans can trust this technology - although the connection to trust is not always made explicit, per se.

Today's artificial assistants are typically prompted to perform tasks through direct, imperative commands such as \emph{Set a timer} or \emph{Pick up the box}. However, to progress toward more natural exchanges between humans and these assistants, it is important to understand the way non-imperative utterances can indirectly elicit action of an addressee. In this paper, we investigate command types in the setting of a grounded, collaborative game. We focus on a less understood family of utterances for eliciting agent action, locatives like \emph{The chair is in the other room}, and demonstrate how these utterances indirectly command in specific game state contexts. Our work shows that models with domain-specific grounding can effectively realize the pragmatic reasoning that is necessary for more robust natural language interaction.

Several weeks ago I wrote a post on Bayesian statistics. I was very interested in the implementation of Bayesian statistics, especially for complex problems which are more easily solved with simulation rather than mathematical manipulation. I wrote the article with a specific audience in mind: namely those that knew the basics of Bayesian statistics, but had no idea how to implement it. As an astute commenter pointed out, in my excitement to implement my Bayesian program, I skimmed over several key points of Bayesian statistics and woefully mis-represented others. Let's fix that now, shall we! Let's talk about the basics of Bayesian statistics, and then move up to simulating them.

Trading is a competitive business. You need great people and great technology, of course, but also trading strategies that make money. Where do those strategies come from? In this post we'll discuss how the interplay of data, math and technology informs how we develop and run strategies. Machine learning (ML) at Jane Street begins, unsurprisingly, with data.

XGBoost is a library that is designed for boosted (tree) algorithms. It has become a popular machine learning framework among data science practitioners, especially on Kaggle, which is a platform for data prediction competitions where researchers post their data and statisticians and data miners compete to produce the best models. For structured learning problems on Kaggle, it can be difficult to get into the top 10 without including XGBoost. Typically, data scientists use multi-thread single machines to train XGBoost models. Very few people have deployed XGBoost on a distributed environment and achieved good performance.

Detecting and segmenting salient objects in natural scenes, often referred to as salient object detection, has attracted a lot of interest in computer vision. While many models have been proposed and several applications have emerged, yet a deep understanding of achievements and issues is lacking. We aim to provide a comprehensive review of the recent progress in salient object detection and situate this field among other closely related areas such as generic scene segmentation, object proposal generation, and saliency for fixation prediction. Covering 228 publications, we survey i) roots, key concepts, and tasks, ii) core techniques and main modeling trends, and iii) datasets and evaluation metrics in salient object detection. We also discuss open problems such as evaluation metrics and dataset bias in model performance and suggest future research directions.

As we can see, the statsmodels library allows us to generate highly detailed output on a level similar to R, with additional statistics such as skew, kurtosis, R-Squared and AIC. While these readings can be generated through scipy or sklearn, doing so is a more intensive process and in many cases these statistics must be calculated individually.

Many studies in biomedical and health sciences involve small sample sizes due to logistic or financial constraints. Often, identifying weak (but scientifically interesting) associations between a set of predictors and a response necessitates pooling datasets from multiple diverse labs or groups. While there is a rich literature in statistical machine learning to address distributional shifts and inference in multi-site datasets, it is less clear ${\it when}$ such pooling is guaranteed to help (and when it does not) -- independent of the inference algorithms we use. In this paper, we present a hypothesis test to answer this question, both for classical and high dimensional linear regression. We precisely identify regimes where pooling datasets across multiple sites is sensible, and how such policy decisions can be made via simple checks executable on each site before any data transfer ever happens. With a focus on Alzheimer's disease studies, we present empirical results showing that in regimes suggested by our analysis, pooling a local dataset with data from an international study improves power.