Understanding and reasoning about places and their relationships are critical for many applications. Places are traditionally curated by a small group of people as place gazetteers and are represented by an ID with spatial extent, category, and other descriptions. However, a place context is described to a large extent by movements made from/to other places. Places are linked and related to each other by these movements. This important context is missing from the traditional representation. We present DeepMove, a novel approach for learning latent representations of places. DeepMove advances the current deep learning based place representations by directly model movements between places. We demonstrate DeepMove's latent representations on place categorization and clustering tasks on large place and movement datasets with respect to important parameters. Our results show that DeepMove outperforms state-of-the-art baselines. DeepMove's representations can provide up to 15% higher than competing methods in matching rate of place category and result in up to 39% higher silhouette coefficient value for place clusters. DeepMove is spatial and temporal context aware. It is scalable. It outperforms competing models using much smaller training dataset (a month or 1/12 of data). These qualities make it suitable for a broad class of real-world applications.

Although sometimes default values exist, there is no agreed upon principle for their definition (but see our recent work in in (Probst et al., 2018) for a potential approach). Automatic tuning of such parameters is a possible solution (Claesen and Moor, 2015), but comes with a considerable computational burden. Meta-learning tries to decrease this cost (Feurer et al., 2015), by reusing information of previous runs of the algorithm on similar datasets, which obviously requires access to such prior empirical results. With this paper we provide a freely accessible meta dataset that contains around 2.5 million runs of six different machine learning algorithms on 38 classification datasets. Large, freely available datasets like Imagenet (Deng et al., 2009) are important for the progress of machine learning, we hope to support developments in the area of meta-learning and benchmarking, meta-learning and hyperparameter tuning with our work here. While similar meta-datasets have been created in the past, we were not able to access them by the links provided in their respective papers: Smith et al. (2014) provides a repository with Weka-based machine learning experiments on 72 data sets, 9 machine learning algorithms, 10 hyperparameter settings for each algorithm, and several meta-features of each data set. Reif (2012) created a meta-dataset based on machine learning experiments on 83 datasets, 6 classification algorithms, and 49 meta features. In this paper, we describe our experimental setup, to specify how our meta-dataset is created by running random machine learning experiments through the OpenML platform (Vanschoren et al., 2013) and how to access our results.

In this work, we propose a novel framework for privacy-preserving client-distributed machine learning. It is motivated by the desire to achieve differential privacy guarantees in the local model of privacy in a way that satisfies all systems constraints using asynchronous client-server communication and provides attractive model learning properties. We call it "Draw and Discard" because it relies on random sampling of models for load distribution (scalability), which also provides additional server-side privacy protections and improved model quality through averaging. We present the mechanics of client and server components of "Draw and Discard" and demonstrate how the framework can be applied to learning Generalized Linear models. We then analyze the privacy guarantees provided by our approach against several types of adversaries and showcase experimental results that provide evidence for the framework's viability in practical deployments.

We propose a novel end-to-end neural network architecture that, once trained, directly outputs a probabilistic clustering of a batch of input examples in one pass. It estimates a distribution over the number of clusters $k$, and for each $1 \leq k \leq k_\mathrm{max}$, a distribution over the individual cluster assignment for each data point. The network is trained in advance in a supervised fashion on separate data to learn grouping by any perceptual similarity criterion based on pairwise labels (same/different group). It can then be applied to different data containing different groups. We demonstrate promising performance on high-dimensional data like images (COIL-100) and speech (TIMIT). We call this ``learning to cluster'' and show its conceptual difference to deep metric learning, semi-supervise clustering and other related approaches while having the advantage of performing learnable clustering fully end-to-end.

Increasing numbers of software vulnerabilities are discovered every year whether they are reported publicly or discovered internally in proprietary code. These vulnerabilities can pose serious risk of exploit and result in system compromise, information leaks, or denial of service. We leveraged the wealth of C and C++ open-source code available to develop a large-scale function-level vulnerability detection system using machine learning. To supplement existing labeled vulnerability datasets, we compiled a vast dataset of millions of open-source functions and labeled it with carefully-selected findings from three different static analyzers that indicate potential exploits. Using these datasets, we developed a fast and scalable vulnerability detection tool based on deep feature representation learning that directly interprets lexed source code. We evaluated our tool on code from both real software packages and the NIST SATE IV benchmark dataset. Our results demonstrate that deep feature representation learning on source code is a promising approach for automated software vulnerability detection.

It is now commonplace for researchers across a variety of fields to fit machine learning models on complex data to make predictions. The complexity of these data (e.g., large number of features, nonlinear relationships with the response) often means it is difficult to determine a priori what machine learning modeling routine and what descriptors (also known as features, predictors, or covariates) will result in the best performance. A common approach to this problem is to fit many descriptor set and modeling routine (DM) combinations, and then compute measures of prediction performance for held out data to choose a DM combination by assessing relative performance. Often in a particular domain, there are only a few modeling routines that are widely accepted, and researchers tend to use these methods exclusively. Unfortunately, this will not always work well for every data set and researchers might learn from other fields where different modeling methods tend to be more succesful. There are a myraid of modeling methods implemented in R that may be worthwhile for researchers to try (see Hastie et al. (2009) and Kuhn and Johnson (2013) for an overview of these methods). However, the lack of knowledge of the syntactic minutiae and statistical methodology that is required to fit and compare different modeling routines in R often prohibits users from attempting them.

We consider the optimization of an uncertain objective over continuous and multi-dimensional decision spaces in problems in which we are only provided with observational data. We propose a novel algorithmic framework that is tractable, asymptotically consistent, and superior to comparable methods on example problems. Our approach leverages predictive machine learning methods and incorporates information on the uncertainty of the predicted outcomes for the purpose of prescribing decisions. We demonstrate the efficacy of our method on examples involving both synthetic and real data sets.

The problem of distributed representation learning is one in which multiple sources of information $X_1,\ldots,X_K$ are processed separately so as to learn as much information as possible about some ground truth $Y$. We investigate this problem from information-theoretic grounds, through a generalization of Tishby's centralized Information Bottleneck (IB) method to the distributed setting. Specifically, $K$ encoders, $K \geq 2$, compress their observations $X_1,\ldots,X_K$ separately in a manner such that, collectively, the produced representations preserve as much information as possible about $Y$. We study both discrete memoryless (DM) and memoryless vector Gaussian data models. For the discrete model, we establish a single-letter characterization of the optimal tradeoff between complexity (or rate) and relevance (or information) for a class of memoryless sources (the observations $X_1,\ldots,X_K$ being conditionally independent given $Y$). For the vector Gaussian model, we provide an explicit characterization of the optimal complexity-relevance tradeoff. Furthermore, we develop a variational bound on the complexity-relevance tradeoff which generalizes the evidence lower bound (ELBO) to the distributed setting. We also provide two algorithms that allow to compute this bound: i) a Blahut-Arimoto type iterative algorithm which enables to compute optimal complexity-relevance encoding mappings by iterating over a set of self-consistent equations, and ii) a variational inference type algorithm in which the encoding mappings are parametrized by neural networks and the bound approximated by Markov sampling and optimized with stochastic gradient descent. Numerical results on synthetic and real datasets are provided to support the efficiency of the approaches and algorithms developed in this paper.

We examine the supervised learning problem in its continuous setting and give a general optimality condition through techniques of functional analysis and the calculus of variations. This enables us to solve the optimality condition for the desired function u numerically and make several comparisons with other widely utilized supervised learning models. We employ the accuracy and area under the receiver operating characteristic curve as metrics of the performance. Finally, 3 analyses are conducted based on these two mentioned metrics where we compare the models and make conclusions to determine whether or not our method is competitive.

Generative Adversarial Networks are powerful generative models that are able to model the manifold of natural images. We leverage this property to perform manifold regularization by approximating a variant of the Laplacian norm using a Monte Carlo approximation that is easily computed with the GAN. When incorporated into the semi-supervised feature-matching GAN we achieve state-of-the-art results for GAN-based semi-supervised learning on CIFAR-10 and SVHN benchmarks, with a method that is significantly easier to implement than competing methods. We also find that manifold regularization improves the quality of generated images, and is affected by the quality of the GAN used to approximate the regularizer.