"We consider problems of Bayesian inference for a spatial epidemic on a graph, where the final state of the epidemic corresponds to bond percolation, and where only the set or number of finally infected sites is observed. We develop appropriate Markov chain Monte Carlo algorithms, demonstrating their effectiveness, and we study problems of optimal experimental design. In particular, we demonstrate that for lattice-based processes an experiment on a sparsified lattice can yield more information on model parameters than one conducted on a complete lattice. We also prove some probabilistic results about the behaviour of estimators associated with large infected clusters."

"We engineer an algorithm to solve the approximate dictionary matching problem. Given a list of words $\mathcal{W}$, maximum distance $d$ fixed at preprocessing time and a query word $q$, we would like to retrieve all words from $\mathcal{W}$ that can be transformed into $q$ with $d$ or less edit operations. We present data structures that support fault tolerant queries by generating an index. On top of that, we present a generalization of the method that eases memory consumption and preprocessing time significantly. At the same time, running times of queries are virtually unaffected. We are able to match in lists of hundreds of thousands of words and beyond within microseconds for reasonable distances."

"The effects of several nonlinear regularization techniques are discussed in the framework of 3D seismic tomography. Traditional, linear, $\ell_2$ penalties are compared to so-called sparsity promoting $\ell_1$ and $\ell_0$ penalties, and a total variation penalty. Which of these algorithms is judged optimal depends on the specific requirements of the scientific experiment. If the correct reproduction of model amplitudes is important, classical damping towards a smooth model using an $\ell_2$ norm works almost as well as minimizing the total variation but is much more efficient. If gradients (edges of anomalies) should be resolved with a minimum of distortion, we prefer $\ell_1$ damping of Daubechies-4 wavelet coefficients.…"

"Assume that we observe a large number of curves, all of them with identical, although unknown, shape, but with a different random shift. The objective is to estimate the individual time shifts and their distribution. Such an objective appears in several biological applications like neuroscience or ECG signal processing, in which the estimation of the distribution of the elapsed time between repetitive pulses with a possibly low signal-noise ratio, and without a knowledge of the pulse shape is of interest. We suggest an M-estimator leading to a three-stage algorithm: we split our data set in blocks, on which the estimation of the shifts is done by minimizing a cost criterion based on a functional of the periodogram; the estimated shifts are then plugged into a standard density estimator. We show that under mild regularity assumptions the density estimate converges weakly to the true shift distribution. The theory is applied both to simulations and to alignment of real ECG signals.…"

"To define oscillatory movements of securities market, we put in the non-local extension of Ito- equation for wavelet-images of random processes. It is proposed an algorithm of creation of evolutionary equation and a model of prediction of the most probable price movement path. It is carried out experimental validation of findings."

"This paper presents a novel type-2 Fuzzy logic System to define the Shape of a facial component with the crisp output. This work is the part of our main research effort to design a system (called FASY) which offers a novel face construction approach based on the textual description and also extracts and analyzes the facial components from a face image by an efficient technique. The Fuzzy model, designed in this paper, takes crisp value of width and height of a facial component and produces the crisp value of Shape for different facial components. This method is designed using Matlab 6.5 and Visual Basic 6.0 and tested with the facial components extracted from 200 male and female face images of different ages from different face databases."

"…We tested our algorithm on a small benchmark graph and on a network of about 500 papers in information science (weighted with the Salton index of bibliographic coupling). In our tests, this approach results in characteristic ranges of resolution where a large resolution change does not lead to a growth of the natural community. Such stable modules were also obtained by applying the LFK algorithm but since we determine communities for all resolution values in one run, our approach is faster than the LFK reference. And our algorithm reveals the hierarchical structure of the graph more easily."

"We introduce a simple criterion, the CAR score, for ranking and selecting variables in linear regression. The CAR score arises naturally in the best predictor formulation of the linear model, offers a canonical decomposition of the proportion of explained variance, and also takes account of correlation and grouping structure among explanatory variables. As population quantity the CAR score is not tied to any specific inference paradigm. Variable selection based on AIC, $C_p$, BIC, and other information criteria is shown to be equivalent to thresholding CAR scores at a fixed level, whereas using false discovery rates corresponds to an adaptive cutoff. In computer simulations we show that CAR scores are highly effective for variable selection with a prediction error that compares favorable with the elastic net and similar regression procedures. We illustrate the approach by analyzing diabetes data as well as gene expression data from the human frontal cortex."

"A single target is hidden at a location chosen from a predetermined probability distribution. Then, a searcher must find a second probability distribution from which random search points are sampled such that the target is found in the minimum number of trials. Here it will be shown that if the searcher must get very close to the target to find it, then the best search distribution is proportional to the square root of the target distribution…"

"A popular method for selecting the number of clusters is based on stability arguments: one chooses the number of clusters such that the corresponding clustering results are "most stable". In recent years, a series of papers has analyzed the behavior of this method from a theoretical point of view. However, the results are very technical and difficult to interpret for non-experts. In this paper we give a high-level overview about the existing literature on clustering stability. In addition to presenting the results in a slightly informal but accessible way, we relate them to each other and discuss their different implications."

"Systems whose organization displays causal asymmetry constraints, from evolutionary trees to river basins or transport networks, can be often described in terms of directed paths (causal flows) on a discrete state space. Such a set of paths defines a feed-forward, acyclic network. A key problem associated with these systems involves characterizing their intrinsic degree of path reversibility: given an end node in the graph, what is the uncertainty of recovering the process backwards until the origin? Here we propose a novel concept, \textit{topological reversibility}, which rigorously weigths such uncertainty in path dependency quantified as the minimum amount of information required to successfully revert a causal path.…"

"The Lin-Kernighan heuristic is known to be one of the most successful heuristics for the Traveling Salesman Problem (TSP). It has also proven its efficiency in application to some other problems. In this paper we discuss possible adaptations of TSP heuristics for the Generalized Traveling Salesman Problem (GTSP) and focus on the case of the Lin-Kernighan algorithm. At first, we provide an easy-to-understand description of the original Lin-Kernighan heuristic. Then we propose several adaptations, both trivial and complicated. Finally, we conduct a fair competition between all the variations of the Lin-Kernighan adaptation and some other GTSP heuristics. It appears that our adaptation of the Lin-Kernighan algorithm for the GTSP reproduces the success of the original heuristic. Different variations of our adaptation outperform all other heuristics in a wide range of trade-offs between solution quality and running time, making Lin-Kernighan the state-of-the-art GTSP local search."

"Recovery of the sparsity pattern (or support) of a sparse vector from a small number of noisy linear samples is a common problem that arises in signal processing and statistics. In the high dimensional setting, it is known that recovery with a vanishing fraction of errors is impossible if the sampling rate and per-sample signal-to-noise ratio (SNR) are finite constants independent of the length of the vector. In this paper, it is shown that recovery with an arbitrarily small but constant fraction of errors is, however, possible, and that in some cases a computationally simple thresholding estimator is near-optimal.…"

"The beam-splitting method for fine-heterogeneity cor- rection will inevitably multiply beams to transport and thus will slow down dose calculation. With the GDS al- gorithm, the dose convolution is made only once after all the beams have been transported, which minimizes the impact of the beam multiplication on computing time. In fact, for the beams individually split into several tens, the calculation time increased only by several times with the GDS. This algorithmic framework will thus enable fast and accurate treatment planning of heavy charged particle ra- diotherapy in the presence of density heterogeneity finer than the size of intrinsic beam blurring."

"In this paper, the performances of the quasi-Newton BFGS algorithm, the NEWUOA derivative free optimizer, the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), the Differential Evolution (DE) algorithm and Particle Swarm Optimizers (PSO) are compared experimentally on benchmark functions reflecting important challenges encountered in real-world optimization problems. Dependence of the performances in the conditioning of the problem and rotational invariance of the algorithms are in particular investigated."

"…Since the produced mesh is applied to the meshless-type methods, the connectivity of the points is not used and only the grid points are important, though the grid lines are utilized in the adapting process. The performance of the adaptive points is exam- ined by considering a collocation meshless method which is based on interpolation in terms of a set of radial basis functions. … Some experimental results will be presented to illustrate the effectiveness of the proposed method."

"We present an efficient solution to the following problem, of relevance in a numerical optimization scheme: calculation of integrals of the type \[\iint_{T \cap \{f\ge0\}} \phi_1\phi_2 \, dx\,dy\] for quadratic polynomials $f,\phi_1,\phi_2$ on a plane triangle $T$. The naive approach would involve consideration of the many possible shapes of $T\cap\{f\geq0\}$ (possibly after a convenient transformation) and parameterizing its border, in order to integrate the variables separately. Our solution involves partitioning the triangle into smaller triangles on which integration is much simpler."

"An easy game: Only paper and pencil needed. A game for one player.
And an open game: Will you improve one of the current world records???"

"The study of high-dimensional differential equations is challenging and difficult due to the analytical and computational intractability. Here, we significantly improve the speed of waveform relaxation (WR), a method to simulate high-dimensional differential-algebraic equations. This new method termed adaptive waveform relaxation (AWR) is tested on a communication network example. Further we analyze different heuristics for computing graph partitions tailored to adaptive waveform relaxation."

"The key insight is that a single strong heuristic will do worse than a collective of individually weak but diverse heuristics. The problems are too hard for any one heuristic to solve perfectly, but the diverse heuristics can, so to speak, cover each others' weaknesses and help each other out when they get stuck; a single strong heuristic can't. A collection of diverse strong heuristics would be even better, but the strong heuristics for a problem tend to be similar to each other, so a group of them lacks diversity. In problem solving and prediction, diversity is exactly as important as individual ability."

To read in context of current practices of Pareto-GP model discovery: are there any cultural similarities <i>at all</i> between these people and the GP practitioners' approach?

via Nic McPhee

"In recent research experiments, Derek Rucker and Adam Galinsky, found that people who felt powerless were willing to pay more money for luxury or status items than people who’d been conditioned to feel more powerful and in control."

"The following principles are fundamental to the design and implementation of effective interfaces, whether for traditional GUI environments or the web. Of late, many web applications have reflected a lack of understanding of many of these principles of interaction design, to their great detriment. Because an application or service appears on the web, the principles do not change. If anything, applying these principles become even more important."

"These are ten general principles for user interface design. They are called "heuristics" because they are more in the nature of rules of thumb than specific usability guidelines."

"1) Rational solutions need not be universal, instead they will be only effective in appropriate contexts.

2) Rationally acceptable conclusions do not have to follow necessarily from the information given, as acceptance is not to be understood in terms of a formal relationship between propositions but in terms of a practical commitment.

3) The rationality of the conclusions is not necessarily determined by whether they conform to the appropriate rules, indeed the primary focus is removed from conclusions and placed upon actions that are taken on the basis of beliefs and methods that are all subject to further criticism and development. "

"Also, I approach the issue from the direction of looking at the nature of rationality - superstition being perhaps the most infamous example of what is seen as humans failing to be rational. Given such an approach, superstition is of interest as it shows something of how human reasoning works by showing how it fails to work. In particular, it is of interest to me as I think that rationality, not just human rationality but any rationality at all, must be understood to be inherently limited or, to use Herbert Simon’s term, bounded. In that context, the study of superstition becomes the study of the bounds of reason."

"So, in summary– a drop in classification error on test data from .941 to .078. Thats a 17% drop. (Or a 21% drop, depending upon which rate you use as a base.) This from a method that you can implement in basically zero extra work if you already have a linear classifier. Seems worth a try."

Interesting prospect for a Nudge application

"The tendency is worrying, because the implication of bestness and a one true way is closing your options for applying a portfolio of responses for a portfolio of needs – which is typically what any complex human systems intervention like knowledge management needs. Knowledge management is most of the time about juggling a number of interventions in a shifting ecology of needs. There are no simple recipes and there are no single best approaches."

Subjective randomness.

"When the mind fixates on absolute discontinuities, mischief is often in the offing..."

"Best guess beats careful planning"

Because they're on my mind, I have to wonder about the implications for agile software methodologies, waterfall project planning, and the desire we have to know what's happening ahead of time. People who are "poor planners" and "procrastinators" often str

First-hand account of the Beer Game simulation.