ANDREW LEAVER-FAY: Capturing Atomic Interactions with a Graphical Framework in Computational Protein Design. (Under the direction of Jack Snoeyink) A protein’s amino acid sequence determines both its chemical and its physical structures, and together these two structures determine its function. Protein designers seek new amino acid sequences with chemical and physical structures capable of perf...

This paper presents the LTAG Workbench, a set of graphical tools and parsers freely available for LTAG. The system can be view as a modern alternative to the XTAG system. We present rst the outlines of the workbench including diierent graphical editors and two chart parsers. The encoding of resources and results is based on an XML application called TagML. We present then future works dedicated...

We present the first general purpose framework for marginal maximum a posteriori estimation of probabilistic program variables. By using a series of code transformations, the evidence of any probabilistic program, and therefore of any graphical model, can be optimized with respect to an arbitrary subset of its sampled variables. To carry out this optimization, we develop the first Bayesian opti...

This paper presents the use of graphical models and copula functions in Estimation of Distribution Algorithms (EDAs) for solving multivariate optimization problems. It is shown in this work how the incorporation of copula functions and graphical models for modeling the dependencies among variables provides some theoretical advantages over traditional EDAs. By means of copula functions and two w...

We propose a Decision-Guided Energy Investment (DGEI) Framework to optimize power, heating, and cooling capacity. The DGEI framework is designed to support energy managers to (1) use the analytical and graphical methodology to determine the best investment option that satisfies the designed evaluation parameters, such as return on investment (ROI) and greenhouse gas (GHG) emissions; (2) develop...

Collective classification predicts class labels simultaneously for a group of related instances, rather than predicting a class for each instance separately. The existing collective classification methods are usually expensive due to the iterative inference in graphical models and their learning procedures based on iterative optimization. When the dataset is large, the cost of maintaining large...

Usability and guessability are two conflicting criteria in assessing the suitability of an image to be used as password in the recognition based graphical authentication systems (RGBSs). We present the first work in this area that uses a new approach, which effectively integrates a series of techniques in order to rank images taking into account the values obtained for each of the dimensions of...