In conventional supervised pattern recognition tasks, model selection is typically accomplished by minimizing the classification error rate on a set of so-called development data, subject to ground-truth labeling by human experts or some other means. In the context of speech processing systems and other large-scale practical applications, however, such labeled development data are typically costly and difficult to obtain. This article proposes an alternative semi-supervised framework for likelihood-based model selection that leverages unlabeled data by using trained classifiers representing each model to automatically generate putative labels. The errors that result from this automatic labeling are shown to be amenable to results from robust statistics, which in turn provide for minimax-optimal censored likelihood ratio tests that recover the nonparametric sign test as a limiting case. This approach is then validated experimentally using a state-of-the-art automatic speech recognition system to select between candidate word pronunciations using unlabeled speech data that only potentially contain instances of the words under test. Results provide supporting evidence for the utility of this approach, and suggest that it may also find use in other applications of machine learning.

The Sample Compression Conjecture of Littlestone & Warmuth has remained unsolved for over two decades. This paper presents a systematic geometric investigation of the compression of finite maximum concept classes. Simple arrangements of hyperplanes in Hyperbolic space, and Piecewise-Linear hyperplane arrangements, are shown to represent maximum classes, generalizing the corresponding Euclidean result. A main result is that PL arrangements can be swept by a moving hyperplane to unlabeled d-compress any finite maximum class, forming a peeling scheme as conjectured by Kuzmin & Warmuth. A corollary is that some d-maximal classes cannot be embedded into any maximum class of VC dimension d+k, for any constant k. The construction of the PL sweeping involves Pachner moves on the one-inclusion graph, corresponding to moves of a hyperplane across the intersection of d other hyperplanes. This extends the well known Pachner moves for triangulations to cubical complexes.

Computability logic (CL) (see http://www.cis.upenn.edu/~giorgi/cl.html) is a recently launched program for redeveloping logic as a formal theory of computability, as opposed to the formal theory of truth that logic has more traditionally been. Formulas in it represent computational problems, "truth" means existence of an algorithmic solution, and proofs encode such solutions. Within the line of research devoted to finding axiomatizations for ever more expressive fragments of CL, the present paper introduces a new deductive system CL12 and proves its soundness and completeness with respect to the semantics of CL. Conservatively extending classical predicate calculus and offering considerable additional expressive and deductive power, CL12 presents a reasonable, computationally meaningful, constructive alternative to classical logic as a basis for applied theories. To obtain a model example of such theories, this paper rebuilds the traditional, classical-logic-based Peano arithmetic into a computability-logic-based counterpart. Among the purposes of the present contribution is to provide a starting point for what, as the author wishes to hope, might become a new line of research with a potential of interesting findings -- an exploration of the presumably quite unusual metatheory of CL-based arithmetic and other CL-based applied systems.

In sports competitions, teams can manipulate the result by, for instance, throwing games. We show that we can decide how to manipulate round robin and cup competitions, two of the most popular types of sporting competitions in polynomial time. In addition, we show that finding the minimal number of games that need to be thrown to manipulate the result can also be determined in polynomial time. Finally, we show that there are several different variations of standard cup competitions where manipulation remains polynomial.

Ensemble methods, such as stacking, are designed to boost predictive accuracy by blending the predictions of multiple machine learning models. Recent work has shown that the use of meta-features, additional inputs describing each example in a dataset, can boost the performance of ensemble methods, but the greatest reported gains have come from nonlinear procedures requiring significant tuning and training time. Here, we present a linear technique, Feature-Weighted Linear Stacking (FWLS), that incorporates meta-features for improved accuracy while retaining the well-known virtues of linear regression regarding speed, stability, and interpretability. FWLS combines model predictions linearly using coefficients that are themselves linear functions of meta-features. This technique was a key facet of the solution of the second place team in the recently concluded Netflix Prize competition. Significant increases in accuracy over standard linear stacking are demonstrated on the Netflix Prize collaborative filtering dataset.

The relationship between equilibrium convergence to a uniform quality distribution and price is investigated in the Q-model, a self-organizing, evolutionary computational matching model of a fixed-price post-secondary higher education created by Ortmann and Slobodyan (2006). The Q-model is replicated with price equaling 100% its Ortmann and Slobodyan (2006) value, Varying the fixed price between 0% and 200% reveals thresholds at which the Q-model reaches different market clustering configurations. Results indicate structural market robustness to prices less than 100% and high sensitivity to prices greater than 100%.

PMRFSim is a proof of concept geospatial social agent-based simulation capable of examining the interactions of 60,000+ agents over a simulated year within a few minutes. PMRFSim utilizes real world data from sources ranging from the U.S. Census Bureau, a regional sociologist, and base security. PMRFSim models two types of agents, normal and adverse agents. Adverse agents have harmful intent and goals to spread negative sentiment and acquire intelligence. All agents are endowed with demographic and geospatial attributes. Agents interact with each other and respond to events. PMRFSim allows an analyst to construct various what-if scenarios and generates numerous graphs that characterize the social landscape. This analysis is intended to aid public affairs officers understand the social landscape.

Multi-representational architectures exploit diversity to yield the breadth of capabilities required for intelligent behavior in the world, but in so doing can sacrifice too much of the complementary benefits of architectural uniformity. The proposal here is to couple the benefits of diversity and uniformity through establishment of a uniform graph-based implementation level for diverse architectures.

The availability of online documents coupled with emergent text mining methods has opened new research horizons. To achieve their potential, mining technologies need to be theoretically focused. We present data theory as a crucial component of text mining, and provide a substantive proto- theory from the synthesis of complex multigames, prototype concepts, and emotio-cognitive orientation fields. We discuss how the data theory presented informs the application of text mining to mining discourse(s) and how, in turn, this allows for modeling across contextual thresholds. Finally, the relationship between discourse mining, data theory, and thresholds is illustrated with an historical example, the events surrounding the 1992 civil war in Tajikistan.

Creating new concepts from data is a hard problem in the development of cognitive architectures, but one that must be solved for the BICA community to declare success.  Two concept generation algorithms are presented here that are appropriate to different levels of concept abstraction: state-space partitioning with decision trees and context-based similarity.