Identifying small subsets of features that are relevant for prediction and/or classification tasks is a central problem in machine learning and statistics. The feature selection task is especially important, and computationally difficult, for modern datasets where the number of features can be comparable to, or even exceed, the number of samples. Here, we show that feature selection with Bayesian inference takes a universal form and reduces to calculating the magnetizations of an Ising model, under some mild conditions. Our results exploit the observation that the evidence takes a universal form for strongly-regularizing priors --- priors that have a large effect on the posterior probability even in the infinite data limit. We derive explicit expressions for feature selection for generalized linear models, a large class of statistical techniques that include linear and logistic regression. We illustrate the power of our approach by analyzing feature selection in a logistic regression-based classifier trained to distinguish between the letters B and D in the notMNIST dataset.

State-space models have been successfully used for more than fifty years in different areas of science and engineering. We present a procedure for efficient varia-tional Bayesian learning of nonlinear state-space models based on sparse Gaussian processes. The result of learning is a tractable posterior over nonlinear dynamical systems. In comparison to conventional parametric models, we offer the possibility to straightforwardly trade off model capacity and computational cost whilst avoiding overfitting. Our main algorithm uses a hybrid inference approach combining variational Bayes and sequential Monte Carlo.

The term "generalized" refers to the fact that problem (1) reduces to the standard lasso problem (Tibshirani 1996, Chen et al. 1998) when D I, but yields different problems with different choices of the penalty matrix D. We will assume that X has full column rank (i.e., rank(X) p), so as to ensure a unique solution in (1) for all values of λ. Our main contribution is to derive efficient implementations of the generalized lasso dual path algorithm of Tibshirani & Taylor (2011). This algorithm computes the solution ˆβ(λ) in (1) over the full range of regularization parameter values λ [0,). We present an efficient implementation for a general penalty matrix D, as well as specialized, extra-efficient implementations for two special classes of generalized lasso problems: fused lasso and trend filtering problems. The algorithms that we describe in this work are all implemented in the genlasso R package, freely available on the CRAN repository (R Development Core Team 2008). We note that the fused lasso and trend filtering problems are known, well-established problems (early references for fused lasso are Land & Friedman (1996), Tibshirani et al. (2005), and early works on trend filtering are Steidl et al. (2006), Kim et al. (2009)). These problems are not original to the generalized lasso framework, but the latter framework simply provides a useful, unifying perspective from which we can study them. We give a brief overview here; see the aforementioned references for more discussion, or Section 2 of Tibshirani & Taylor (2011), and also Section 6 of this paper, for examples and figures.

Minimizing the number of reshuffling operations at maritime container terminals incorporates the Pre-Marshalling Problem (PMP) as an important problem. Based on an analysis of existing solution approaches we develop new heuristics utilizing specific properties of problem instances of the PMP. We show that the heuristic performance is highly dependent on these properties. We introduce a new method that exploits a greedy heuristic of four stages, where for each of these stages several different heuristics may be applied. Instead of using randomization to improve the performance of the heuristic, we repetitively generate a number of solutions by using a combination of different heuristics for each stage. In doing so, only a small number of solutions is generated for which we intend that they do not have undesirable properties, contrary to the case when simple randomization is used. Our experiments show that such a deterministic algorithm significantly outperforms the original nondeterministic method when the quality of found solutions is observed, with a much lower number of generated solutions.

This paper proposes the use of the Brownian distance correlation to conduct a lead-lag analysis of financial and economic time series. When this methodology is applied to asset prices, the non-linear relationships identified may improve the price discovery process of these assets. The Brownian distance correlation determines relationships similar to those identified by the linear Granger causality test, and it also uncovers additional non-linear relationships among the log prices of oil, coal, and natural gas.

Film cinematography has been developed and applied for more than a century to involve and engage the viewer in visual storytelling. Interactive storytelling games can benefit from these cinematic conventions to enhance visual experience. However, even conversation scenes in games are highly dynamic, and pre-authoring camera parameters using cinematography principles is often insufficient. This paper proposes an automatic Visual Director System focused on dynamic conversation scenes involving three characters and reports on work in progress on a prototype applied to the recreation of a movie scene. Based on principles of cinematography and the study of film scenes, cinematic conventions for triadic conversations are encoded modularly as an artificial intelligence game component that selects suitable shots for dynamic scenes.

We demonstrate the Chimeria Platform that computationally models aspects of social identity dynamics for use in digital media such as in videogames and social networks. The Engine models users’ degrees of membership across multiple categories as gradient values, enabling more representational nuance than binary statuses of member/nonmember. The Application Interface handles user interaction and visuals for experiencing the narratives. Domain Epistemologies specify domain-specific ontologies that describe cultural knowledge and beliefs for each narrative. Our Visual Narrative Editor GUI is being developed to make authoring more accessible to a wider audience.

In this paper, we translate a model of value-based emo- tional agents into an architecture for narrative characters and we validate it in a narrative scenario. The advantage of using such model is that different moral behaviors can be obtained as a consequence of the emotional ap- praisal of moral values, a desirable feature for digital storytelling techniques.

AI research in interactive narrative often lacks specificity as to the player experience it is trying to enable. In this paper, we consider a set of desirable elements from narrative and interactive experiences, and show by looking at playable experiences from industry and academia that combining them has the potential to be limited or self-defeating. To address these issues, we propose opportunistic storytelling , a set of design principles for near-term playable interactive narratives.

We discuss a design pattern found in expressive storygames, the eurekon, which describes a specific dynamic arising from some adventure game puzzles where the player experiences a moment of revelation connecting the narrative and ludic planes. Eurekons have largely been designed out of modern storygames in favor of patterns that reduce the possibility of failure (as seen in the fall of the "puzzle" and rise of the "quest"), but this shift often eliminates the unique pleasures often found in a successful eurekon. We demonstrate both how the eurekon is a useful concept in analyzing existing adventure games and how it can inform designers hoping to create more successful eurekons.