Exact Combinatorial Optimization with Graph Convolutional Neural Networks

Combinatorial optimization problems are typically tackled by the branch-and-bound paradigm. We propose to learn a variable selection policy for branch-and-bound in mixed-integer linear programming, by imitation learning on a diversified variant of the strong branching expert rule. We encode states as bipartite graphs and parameterize the policy as a graph convolutional neural network. Experiments on a series of synthetic problems demonstrate that our approach produces policies that can improve upon expert-designed branching rules on large problems, and generalize to instances significantly larger than seen during training.

Preharvest factors affecting quality and antioxidant levels on Abate Fétel pears: study of superficial scald with a multivariate statistical approach and neural networks

There are only a few insights concerning the influence that agronomic and management variability may have on superficial scald (SS) in pears. Abate Fétel pears were picked during three seasons (2018, 2019 and 2020) from thirty commercial orchards in the Emilia Romagna region, Italy. Using a multivariate statistical approach, high heterogeneity between farms for SS development after cold storage with regular atmosphere was demonstrated. Indeed, some factors seem to affect SS in all growing seasons: high yields, soil texture, improper irrigation and Nitrogen management, use of plant growth regulators, late harvest, precipitations, Calcium and cow manure, presence of nets, orchard age, training system and rootstock. Afterwards, we explored the spatio/temporal variability of fruit attributes in two pear orchards. Environmental and physiological spatial variables were recorded by a portable RTK GPS. High spatial variability of the SS index was observed. Through a geostatistical approach, some characteristics, including soil electrical conductivity and fruit size, have been shown to be negatively correlated with SS. Moreover, regression tree analyses were applied suggesting the presence of threshold values of antioxidant capacity, total phenolic content, and acidity against SS. High pulp firmness and IAD values before storage, denoting a more immature fruit, appeared to be correlated with low SS. Finally, a convolution neural networks (CNN) was tested to detect SS and the starch pattern index (SPI) in pears for portable device applications. Preliminary statistics showed that the model for SS had low accuracy but good precision, and the CNN for SPI denoted good performances compared to the Ctifl and Laimburg scales. The major conclusion is that Abate Fétel pears can potentially be stored in different cold rooms, according to their origin and quality features, ensuring the best fruit quality for the final consumers. These results might lead to a substantial improvement in the Italian pear industry.

Tree regularized neural networks for key point analysis

This thesis contributes to the ArgMining 2021 shared task on Key Point Analysis. Key Point Analysis entails extracting and calculating the prevalence of a concise list of the most prominent talking points, from an input corpus. These talking points are usually referred to as key points. Key point analysis is divided into two subtasks: Key Point Matching, which involves assigning a matching score to each key point/argument pair, and Key Point Generation, which consists of the generation of key points. The task of Key Point Matching was approached using different models: a pretrained Sentence Transformers model and a tree-constrained Graph Neural Network were tested. The best model was the fine-tuned Sentence Transformers, which achieved a mean Average Precision score of 0.75, ranking 12 compared to other participating teams. The model was then used for the subtask of Key Point Generation using the extractive method in the selection of key point candidates and the model developed for the previous subtask to evaluate them.

Preference Mapping Of Dulce De Leche Commercialized In Brazilian Markets.

Dulce de leche samples available in the Brazilian market were submitted to sensory profiling by quantitative descriptive analysis and acceptance test, as well sensory evaluation using the just-about-right scale and purchase intent. External preference mapping and the ideal sensory characteristics of dulce de leche were determined. The results were also evaluated by principal component analysis, hierarchical cluster analysis, partial least squares regression, artificial neural networks, and logistic regression. Overall, significant product acceptance was related to intermediate scores of the sensory attributes in the descriptive test, and this trend was observed even after consumer segmentation. The results obtained by sensometric techniques showed that optimizing an ideal dulce de leche from the sensory standpoint is a multidimensional process, with necessary adjustments on the appearance, aroma, taste, and texture attributes of the product for better consumer acceptance and purchase. The optimum dulce de leche was characterized by high scores for the attributes sweet taste, caramel taste, brightness, color, and caramel aroma in accordance with the preference mapping findings. In industrial terms, this means changing the parameters used in the thermal treatment and quantitative changes in the ingredients used in formulations.

Inference of gene regulatory networks from time series by Tsallis entropy

Background: The inference of gene regulatory networks (GRNs) from large-scale expression profiles is one of the most challenging problems of Systems Biology nowadays. Many techniques and models have been proposed for this task. However, it is not generally possible to recover the original topology with great accuracy, mainly due to the short time series data in face of the high complexity of the networks and the intrinsic noise of the expression measurements. In order to improve the accuracy of GRNs inference methods based on entropy (mutual information), a new criterion function is here proposed. Results: In this paper we introduce the use of generalized entropy proposed by Tsallis, for the inference of GRNs from time series expression profiles. The inference process is based on a feature selection approach and the conditional entropy is applied as criterion function. In order to assess the proposed methodology, the algorithm is applied to recover the network topology from temporal expressions generated by an artificial gene network (AGN) model as well as from the DREAM challenge. The adopted AGN is based on theoretical models of complex networks and its gene transference function is obtained from random drawing on the set of possible Boolean functions, thus creating its dynamics. On the other hand, DREAM time series data presents variation of network size and its topologies are based on real networks. The dynamics are generated by continuous differential equations with noise and perturbation. By adopting both data sources, it is possible to estimate the average quality of the inference with respect to different network topologies, transfer functions and network sizes. Conclusions: A remarkable improvement of accuracy was observed in the experimental results by reducing the number of false connections in the inferred topology by the non-Shannon entropy. The obtained best free parameter of the Tsallis entropy was on average in the range 2.5 <= q <= 3.5 (hence, subextensive entropy), which opens new perspectives for GRNs inference methods based on information theory and for investigation of the nonextensivity of such networks. The inference algorithm and criterion function proposed here were implemented and included in the DimReduction software, which is freely available at http://sourceforge.net/projects/dimreduction and http://code.google.com/p/dimreduction/.

Neural network model for designing automotive devices using SMD LED

The advantages offered by the electronic component LED (Light Emitting Diode) have resulted in a quick and extensive application of this device in the replacement of incandescent lights. In this combined application, however, the relationship between the design variables and the desired effect or result is very complex and renders it difficult to model using conventional techniques. This paper consists of the development of a technique using artificial neural networks that makes it possible to obtain the luminous intensity values of brake lights using SMD (Surface Mounted Device) LEDs from design data. This technique can be utilized to design any automotive device that uses groups of SMD LEDs. The results of industrial applications using SMD LED are presented to validate the proposed technique.

Gait Pattern Adaptation for an Active Lower-Limb Orthosis Based on Neural Networks

This work deals with neural network (NN)-based gait pattern adaptation algorithms for an active lower-limb orthosis. Stable trajectories with different walking speeds are generated during an optimization process considering the zero-moment point (ZMP) criterion and the inverse dynamic of the orthosis-patient model. Additionally, a set of NNs is used to decrease the time-consuming analytical computation of the model and ZMP. The first NN approximates the inverse dynamics including the ZMP computation, while the second NN works in the optimization procedure, giving an adapted desired trajectory according to orthosis-patient interaction. This trajectory adaptation is added directly to the trajectory generator, also reproduced by a set of NNs. With this strategy, it is possible to adapt the trajectory during the walking cycle in an on-line procedure, instead of changing the trajectory parameter after each step. The dynamic model of the actual exoskeleton, with interaction forces included, is used to generate simulation results. Also, an experimental test is performed with an active ankle-foot orthosis, where the dynamic variables of this joint are replaced in the simulator by actual values provided by the device. It is shown that the final adapted trajectory follows the patient intention of increasing the walking speed, so changing the gait pattern. (C) Koninklijke Brill NV, Leiden, 2011

Comparison between detailed model simulation and artificial neural network for forecasting building energy consumption

There are several ways to attempt to model a building and its heat gains from external sources as well as internal ones in order to evaluate a proper operation, audit retrofit actions, and forecast energy consumption. Different techniques, varying from simple regression to models that are based on physical principles, can be used for simulation. A frequent hypothesis for all these models is that the input variables should be based on realistic data when they are available, otherwise the evaluation of energy consumption might be highly under or over estimated. In this paper, a comparison is made between a simple model based on artificial neural network (ANN) and a model that is based on physical principles (EnergyPlus) as an auditing and predicting tool in order to forecast building energy consumption. The Administration Building of the University of Sao Paulo is used as a case study. The building energy consumption profiles are collected as well as the campus meteorological data. Results show that both models are suitable for energy consumption forecast. Additionally, a parametric analysis is carried out for the considered building on EnergyPlus in order to evaluate the influence of several parameters such as the building profile occupation and weather data on such forecasting. (C) 2008 Elsevier B.V. All rights reserved.

MODELLING AND CONTROL OF ERSW PROCESSES BY NEURONAL NETWORK

A study on the use of artificial intelligence (AI) techniques for the modelling and subsequent control of an electric resistance spot welding process (ERSW) is presented. The ERSW process is characterized by the coupling of thermal, electrical, mechanical, and metallurgical phenomena. For this reason, early attempts to model it using computational methods established as the methods of finite differences, finite element, and finite volumes, ask for simplifications that lead the model obtained far from reality or very costly in terms of computational costs, to be used in a real-time control system. In this sense, the authors have developed an ERSW controller that uses fuzzy logic to adjust the energy transferred to the weld nugget. The proposed control strategies differ in the speed with which it reaches convergence. Moreover, their application for a quality control of spot weld through artificial neural networks (ANN) is discussed.

Relationship between peptide selectivities of human transporters associated with antigen processing and HLA class I molecules

Efficiency of presentation of a peptide epitope by a MHC class I molecule depends on two parameters: its binding to the MHC molecule and its generation by intracellular Ag processing. In contrast to the former parameter, the mechanisms underlying peptide selection in Ag processing are poorly understood. Peptide translocation by the TAP transporter is required for presentation of most epitopes and may modulate peptide supply to MHC class I molecules. To study the role of human TAP for peptide presentation by individual HLA class I molecules, we generated artificial neural networks capable of predicting the affinity of TAP for random sequence 9-mer peptides. Using neural network-based predictions of TAP affinity, we found that peptides eluted from three different HLA class I molecules had higher TAP affinities than control peptides with equal binding affinities for the same HLA class I molecules, suggesting that human TAP may contribute to epitope selection. In simulated TAP binding experiments with 408 HLA class I binding peptides, HLA class I molecules differed significantly with respect to TAP affinities of their ligands, As a result, some class I molecules, especially HLA-B27, may be particularly efficient in presentation of cytosolic peptides with low concentrations, while most class I molecules may predominantly present abundant cytosolic peptides.

Methods for prediction of peptide binding to MHC molecules: A comparative study

Background: A variety of methods for prediction of peptide binding to major histocompatibility complex (MHC) have been proposed. These methods are based on binding motifs, binding matrices, hidden Markov models (HMM), or artificial neural networks (ANN). There has been little prior work on the comparative analysis of these methods. Materials and Methods: We performed a comparison of the performance of six methods applied to the prediction of two human MHC class I molecules, including binding matrices and motifs, ANNs, and HMMs. Results: The selection of the optimal prediction method depends on the amount of available data (the number of peptides of known binding affinity to the MHC molecule of interest), the biases in the data set and the intended purpose of the prediction (screening of a single protein versus mass screening). When little or no peptide data are available, binding motifs are the most useful alternative to random guessing or use of a complete overlapping set of peptides for selection of candidate binders. As the number of known peptide binders increases, binding matrices and HMM become more useful predictors. ANN and HMM are the predictive methods of choice for MHC alleles with more than 100 known binding peptides. Conclusion: The ability of bioinformatic methods to reliably predict MHC binding peptides, and thereby potential T-cell epitopes, has major implications for clinical immunology, particularly in the area of vaccine design.

Virtual models of the HLA class I antigen processing pathway

Antigen recognition by cytotoxic CD8 T cells is dependent upon a number of critical steps in MHC class I antigen processing including proteosomal cleavage, TAP transport into the endoplasmic reticulum, and MHC class 1 binding. Based on extensive experimental data relating to each of these steps there is now the capacity to model individual antigen processing steps with a high degree of accuracy. This paper demonstrates the potential to bring together models of individual antigen processing steps, for example proteosome cleavage, TAP transport, and MHC binding, to build highly informative models of functional pathways. In particular, we demonstrate how an artificial neural network model of TAP transport was used to mine a HLA-binding database so as to identify H LA-binding peptides transported by TAP. This integrated model of antigen processing provided the unique insight that HLA class I alleles apparently constitute two separate classes: those that are TAP-efficient for peptide loading (HLA-B27, -A3, and -A24) and those that are TAP-inefficient (HLA-A2, -B7, and -B8). Hence, using this integrated model we were able to generate novel hypotheses regarding antigen processing, and these hypotheses are now capable of being tested experimentally. This model confirms the feasibility of constructing a virtual immune system, whereby each additional step in antigen processing is incorporated into a single modular model. Accurate models of antigen processing have implications for the study of basic immunology as well as for the design of peptide-based vaccines and other immunotherapies. (C) 2004 Elsevier Inc. All rights reserved.

Computational methods for prediction of T-cell epitopes - a framework for modelling, testing, and applications

Computational models complement laboratory experimentation for efficient identification of MHC-binding peptides and T-cell epitopes. Methods for prediction of MHC-binding peptides include binding motifs, quantitative matrices, artificial neural networks, hidden Markov models, and molecular modelling. Models derived by these methods have been successfully used for prediction of T-cell epitopes in cancer, autoimmunity, infectious disease, and allergy. For maximum benefit, the use of computer models must be treated as experiments analogous to standard laboratory procedures and performed according to strict standards. This requires careful selection of data for model building, and adequate testing and validation. A range of web-based databases and MHC-binding prediction programs are available. Although some available prediction programs for particular MHC alleles have reasonable accuracy, there is no guarantee that all models produce good quality predictions. In this article, we present and discuss a framework for modelling, testing, and applications of computational methods used in predictions of T-cell epitopes. (C) 2004 Elsevier Inc. All rights reserved.

Learning the dynamics of embedded clauses

Recent work by Siegelmann has shown that the computational power of recurrent neural networks matches that of Turing Machines. One important implication is that complex language classes (infinite languages with embedded clauses) can be represented in neural networks. Proofs are based on a fractal encoding of states to simulate the memory and operations of stacks. In the present work, it is shown that similar stack-like dynamics can be learned in recurrent neural networks from simple sequence prediction tasks. Two main types of network solutions are found and described qualitatively as dynamical systems: damped oscillation and entangled spiraling around fixed points. The potential and limitations of each solution type are established in terms of generalization on two different context-free languages. Both solution types constitute novel stack implementations - generally in line with Siegelmann's theoretical work - which supply insights into how embedded structures of languages can be handled in analog hardware.