Maurizio Filippone

Aggregation Algorithm Towards Large-Scale Boolean Network Analysis

The analysis of large-scale Boolean network dynamics is of great importance in understanding complex phenomena where systems are characterized by a large number of components. The computational cost to reveal the number of attractors and the period of each attractor increases exponentially as the number of nodes in the networks increases. This paper presents an efficient algorithm to find attractors for medium to large-scale networks. This is achieved by analyzing subnetworks within the network in a way that allows to reveal the attractors of the full network with little computational cost. In particular, for each subnetwork modeled as a Boolean control network, the input-state cycles are found and they are composed to reveal the attractors of the full network. The proposed algorithm reduces the computational cost significantly, especially in finding attractors of short period, or any periods if the aggregation network is acyclic. Also, this paper shows that finding the best acyclic aggregation is equivalent to finding the strongly connected components of the network graph. Finally, the efficiency of the algorithm is demonstrated on two biological systems, namely a T-cell receptor network and an early flower development network.

Information theoretic novelty detection

We present a novel approach to online change detection problems when the training sample size is small. The proposed approach is based on estimating the expected information content of a new data point and allows an accurate control of the false positive rate even for small data sets. In the case of the Gaussian distribution, our approach is analytically tractable and closely related to classical statistical tests. We then propose an approximation scheme to extend our approach to the case of the mixture of Gaussians. We evaluate extensively our approach on synthetic data and on three real benchmark data sets. The experimental validation shows that our method maintains a good overall accuracy, but significantly improves the control over the false positive rate.

Pseudo-Marginal Bayesian Inference for Gaussian Processes

The main challenges that arise when adopting Gaussian process priors in probabilistic modeling are how to carry out exact Bayesian inference and how to account for uncertainty on model parameters when making model-based predictions on out-of-sample data. Using probit regression as an illustrative working example, this paper presents a general and effective methodology based on the pseudo-marginal approach to Markov chain Monte Carlo that efficiently addresses both of these issues. The results presented in this paper show improvements over existing sampling methods to simulate from the posterior distribution over the parameters defining the covariance function of the Gaussian Process prior. This is particularly important as it offers a powerful tool to carry out full Bayesian inference of Gaussian Process based hierarchic statistical models in general. The results also demonstrate that Monte Carlo based integration of all model parameters is actually feasible in this class of models providing a superior quantification of uncertainty in predictions. Extensive comparisons with respect to state-of-the-art probabilistic classifiers confirm this assertion.

Probabilistic prediction of neurological disorders with a statistical assessment of neuroimaging data modalities

For many neurological disorders, prediction of disease state is an important clinical aim. Neuroimaging provides detailed information about brain structure and function from which such predictions may be statistically derived. A multinomial logit model with Gaussian process priors is proposed to: (i) predict disease state based on whole-brain neuroimaging data and (ii) analyze the relative informativeness of different image modalities and brain regions. Advanced Markov chain Monte Carlo methods are employed to perform posterior inference over the model. This paper reports a statistical assessment of multiple neuroimaging modalities applied to the discrimination of three Parkinsonian neurological disorders from one another and healthy controls, showing promising predictive performance of disease states when compared to nonprobabilistic classifiers based on multiple modalities. The statistical analysis also quantifies the relative importance of different neuroimaging measures and brain regions in discriminating between these diseases and suggests that for prediction there is little benefit in acquiring multiple neuroimaging sequences. Finally, the predictive capability of different brain regions is found to be in accordance with the regional pathology of the diseases as reported in the clinical literature.

Automated, High Accuracy Classification of Parkinsonian Disorders: A Pattern Recognition Approach

Progressive supranuclear palsy (PSP), multiple system atrophy (MSA) and idiopathic Parkinson’s disease (IPD) can be clinically indistinguishable, especially in the early stages, despite distinct patterns of molecular pathology. Structural neuroimaging holds promise for providing objective biomarkers for discriminating these diseases at the single subject level but all studies to date have reported incomplete separation of disease groups. In this study, we employed multi-class pattern recognition to assess the value of anatomical patterns derived from a widely available structural neuroimaging sequence for automated classification of these disorders. To achieve this, 17 patients with PSP, 14 with IPD and 19 with MSA were scanned using structural MRI along with 19 healthy controls (HCs). An advanced probabilistic pattern recognition approach was employed to evaluate the diagnostic value of several pre-defined anatomical patterns for discriminating the disorders, including: (i) a subcortical motor network; (ii) each of its component regions and (iii) the whole brain. All disease groups could be discriminated simultaneously with high accuracy using the subcortical motor network. The region providing the most accurate predictions overall was the midbrain/brainstem, which discriminated all disease groups from one another and from HCs. The subcortical network also produced more accurate predictions than the whole brain and all of its constituent regions. PSP was accurately predicted from the midbrain/brainstem, cerebellum and all basal ganglia compartments; MSA from the midbrain/brainstem and cerebellum and IPD from the midbrain/brainstem only. This study demonstrates that automated analysis of structural MRI can accurately predict diagnosis in individual patients with Parkinsonian disorders, and identifies distinct patterns of regional atrophy particularly useful for this process.

A comparative evaluation of stochastic-based inference methods for Gaussian process models

Gaussian Process (GP) models are extensively used in data analysis given their flexible modeling capabilities and interpretability. The fully Bayesian treatment of GP models is analytically intractable, and therefore it is necessary to resort to either deterministic or stochastic approximations. This paper focuses on stochastic-based inference techniques. After discussing the challenges associated with the fully Bayesian treatment of GP models, a number of inference strategies based on Markov chain Monte Carlo methods are presented and rigorously assessed. In particular, strategies based on efficient parameterizations and efficient proposal mechanisms are extensively compared on simulated and real data on the basis of convergence speed, sampling efficiency, and computational cost.

Applying the Possibilistic c-Means Algorithm in Kernel-Induced Spaces

In this paper, we study a kernel extension of the classic possibilistic c-means. In the proposed extension, we implicitly map input patterns into a possibly high-dimensional space by means of positive semidefinite kernels. In this new space, we model the mapped data by means of the possibilistic clustering algorithm. We study in more detail the special case where we model the mapped data using a single cluster only, since it turns out to have many interesting properties. The modeled memberships in kernel-induced spaces yield a modeling of generic shapes in the input space. We analyze in detail the connections to one-class support vector machines and kernel density estimation, thus, suggesting that the proposed algorithm can be used in many scenarios of unsupervised learning. In the experimental part, we analyze the stability and the accuracy of the proposed algorithm on some synthetic and real datasets. The results show high stability and good performances in terms of accuracy.

Dealing with non-metric dissimilarities in fuzzy central clustering algorithms

Clustering is the problem of grouping objects on the basis of a similarity measure among them. Relational clustering methods can be employed when a feature-based representation of the objects is not available, and their description is given in terms of pairwise (dis)similarities. This paper focuses on the relational duals of fuzzy central clustering algorithms, and their application in situations when patterns are represented by means of non-metric pairwise dissimilarities. Symmetrization and shift operations have been proposed to transform the dissimilarities among patterns from non-metric to metric. In this paper, we analyze how four popular fuzzy central clustering algorithms are affected by such transformations. The main contributions include the lack of invariance to shift operations, as well as the invariance to symmetrization. Moreover, we highlight the connections between relational duals of central clustering algorithms and central clustering algorithms in kernel-induced spaces. One among the presented algorithms has never been proposed for non-metric relational clustering, and turns out to be very robust to shift operations.

Possibilistic approach to biclustering: an application to oligonucleotide microarray data analysis

The important research objective of identifying genes with similar behavior with respect to different conditions has recently been tackled with biclustering techniques. In this paper we introduce a new approach to the biclustering problem using the Possibilistic Clustering paradigm. The proposed Possibilistic Biclustering algorithm finds one bicluster at a time, assigning a membership to the bicluster for each gene and for each condition. The biclustering problem, in which one would maximize the size of the bicluster and minimizing the residual, is faced as the optimization of a proper functional. We applied the algorithm to the Yeast database, obtaining fast convergence and good quality solutions. We discuss the effects of parameter tuning and the sensitivity of the method to parameter values. Comparisons with other methods from the literature are also presented.

Predicting Continuous Conflict Perceptionwith Bayesian Gaussian Processes

Conflict is one of the most important phenomena of social life, but it is still largely neglected by the computing community. This work proposes an approach that detects common conversational social signals (loudness, overlapping speech, etc.) and predicts the conflict level perceived by human observers in continuous, non-categorical terms. The proposed regression approach is fully Bayesian and it adopts automatic relevance determination to identify the social signals that influence most the outcome of the prediction. The experiments are performed over the SSPNet Conflict Corpus, a publicly available collection of 1,430 clips extracted from televised political debates (roughly 12 hours of material for 138 subjects in total). The results show that it is possible to achieve a correlation close to 0.8 between actual and predicted conflict perception.