Adriana Birlutiu

1 Decembrie 1918 University | UAB · Department of Computer Science and Engineering

This paper presents the domain generalization methods Multi-Domain Transfer Component Analysis (Multi-TCA) and Multi-Domain Semi-Supervised Transfer Component Analysis (Multi-SSTCA) which are extensions of the domain adaptation method Transfer Component Analysis to multiple domains. Multi-TCA learns a shared subspace by minimizing the dissimilariti...

This paper discusses a kernel ridge regression (KRR) model for motion estimation in radiotherapy. Using KRR, dense internal motion fields are estimated from high-dimensional surrogates without the need for prior dimensionality reduction. We compare the proposed model to a related approach with dimensionality reduction in the form of principal compo...

The combination of spectroscopic measurements and multivariate calibration techniques (chemometrics) has become a state-of-the-art technology for process analytical chemistry. Changes, intended or unintended, in the environmental conditions, the measurement setup or of the measured substance itself can result in a calibration model that is no longe...

This paper investigates domain generalization: How to use knowledge acquired from related domains and apply it to new domains? Transfer Component Analysis (TCA) learns a shared subspace by minimizing the dissimilarities across domains, while maximally preserving the data variance. We propose Multi-TCA, an extension of TCA to multiple domains as wel...

Computational methods for predicting protein-protein interactions are important tools that can complement high-throughput technologies and guide biologists in designing new laboratory experiments. The proteins and the interactions between them can be described by a network which is characterized by several topological properties. Information about...

The reconstruction of protein-protein interaction networks is nowadays an important challenge in systems biology. Computational approaches can address this problem by complementing high-throughput technologies and by helping and guiding biologists in designing new laboratory experiments. The proteins and the interactions between them form a network...

This paper investigates the integration of clinico-pathological and microRNA data for breast cancer relapse prediction. Clinical and pathological data proved to be relevant in making predictions about cancer disease outcome. The most accurate predictive models can be obtained by using clinico-pathological information together with genomic informati...

Early and non-invasive diagnosis and prognosis are important but challenging problems in cancer clinical management.Our results suggest that the most accurate solutions could be developed combining Artificial Intelligence with OMICS technologies. The new i-Biomarker concept and the methodology for developing panels of i-Biomarkers will be presented...

Understanding and eventually controlling the dynamics of microRNA (miRNA) networks via drug treatments are important yet challenging biomedical research issues. Ordinary differential equations (ODEs), the main mathematical framework of dynamical systems theory, even though considered adequate models for such biomedical networks, are difficult to bu...

The aim of this study is to propose a methodology for developing intelligent systems for cancer diagnosis and evaluate it on bladder cancer. Owing to recent advances in high-throughput experiments, large data repositories are now freely available for use. However, the process of extracting information from these data and transforming it into clinic...

This paper presents a framework for optimizing the preference learning pro-cess. In many real-world applications in which preference learning is involved the avail-able training data is scarce and obtaining labeled training data is expensive. Luckily in many of the preference learning situations data is available from multiple subjects. We use the...

In many supervised learning tasks it can be costly or infeasible to obtain objective, reliable labels. We may, however, be able to obtain a large number of subjective, possibly noisy, labels from multiple annotators. Typically, annotators have different levels of expertise (i.e., novice, expert) and there is considerable diagreement among annotator...

Ecological functional genomics, dealing with the responses of organisms to their natural environment is confronted with a complex pattern of variation and a large number of confounding environmental factors. For gene expression studies to provide meaningful information on conditions deviating from normal, a baseline or normal operating range (NOR)...

Presentation at MLCB 2010

We present an EM-algorithm for the problem of learning preferences with semiparametric models derived from Gaussian processes in the context of multi-task learning. We validate our approach on an audiological data set and show that predictive results for sound quality perception of hearing-impaired subjects, in the context of pairwise comparison ex...

In the last decades enormous advances have been made possible for modelling complex (physical) systems by mathematical equations and computer algorithms. To deal with very long running times of such models a promising approach has been to replace them by stochastic approximations based on a few model evaluations. In this paper we focus on the often...

We present an EM-algorithm for the problem of learning user preferences with Gaussian processes in the context of multi-task learning. We validate our approach on an audiological data set and show that predictive results for sound quality perception of normal hearing and hearingimpaired subjects, in the context of pairwise comparison experiments, c...

Rating players in sports competitions based on game results is one example of paired comparison data analysis. Since an exact Bayesian treatment is intractable, several techniques for approximate inference have been proposed in the literature. In this paper we compare several variants of expectation propagation (EP). EP generalizes assumed density...

We propose a new criterion for experimental design in the context of preference learning. This new criterion makes direct use of the data available from a group of subjects for which the preferences were already learned. Furthermore, we show the connections between this criterion and the standard criteria used in experimen-tal design. Empirical res...

The malaria parasite has a direct influence on the immune system of its host, as witnessed by periodic fevers. It is unclear whether the host immune system has a direct influence on parasite gene expression. The parasite is known to secrete proteins that are thought to manipulate the host immune system. An open question is whether the malaria paras...

Medical devices such as hearing aids contain many tunable parameters. The optimal setting of these parameters depends on the patient's preference (utility) function, which is unknown. This raises two questions: (1) how should we optimize the parameters given partial information about the patient's utility? And (2), what questions do we ask to effic...