The Shape Boltzmann Machine: A Strong Model of Object Shape

A good model of object shape is essential in applications such as segmentation, detection, inpainting and graphics. For example, when performing segmentation, local constraints on the shapes can help where object boundaries are noisy or unclear, and global constraints can resolve ambiguities where background clutter looks similar to parts of the objects. In general, the stronger the model of shape, the more performance is improved. In this paper, we use a type of deep Boltzmann machine (Salakhutdinov and Hinton, International Conference on Artificial Intelligence and Statistics, 2009) that we call a Shape Boltzmann Machine (SBM) for the task of modeling foreground/background (binary) and parts-based (categorical) shape images. We show that the SBM characterizes a strong model of shape, in that samples from the model look realistic and it can generalize to generate samples that differ from training examples. We find that the SBM learns distributions that are qualitatively and quantitatively better than existing models for this task.     

A compendium of comparison function results

The use of comparison functions has become standard in systems and control theory, particularly for the purposes of studying stability properties. The use of these functions typically allows elegant and succinct statements of stability properties such as asymptotic stability and input-to-state stability and its several variants. Furthermore, over the last 20 years several inequalities involving these comparison functions have been developed that simplify their manipulation in the service of proving more significant results. Many of these inequalities have appeared in the body of proofs or in appendices of various papers. Our goal herein is to collect these inequalities in one place.     

Biochemical activities during lipid accumulation inCandida curvata

Intracellular and extracellular concentrations of citrate and the specific activities of ten different enzymes in Candida curvata D were examined in relation to lipid biosynthesis in batch and continuous culture. Citrate was found to accumulate prior to lipid production and declined markedly as lipid accumulated in batch culture. The cells excreted citrate as the culture became nitrogen-limiting after 30 hr of growth, but little more was expelled after 40 hr when lipid accumulation was more marked. In continuous culture, only low levels of citrate were detected at the lower dilution rates and citrate was completely absent from both the cells and medium above a dilution rate of 0.1/hr. The activity of malic enzyme, malate dehydrogenase and ATP:citrate lyase increased in batch culture on lipid accumulated and, in continuous culture, both malic enzyme and ATP:citrate lyase varied in parallel with the specific rate of lipid synthesis which increased with increasing dilution rate. Activity of malate dehydrogenase, citrate synthase and glucose-6-phosphate dehydrogenase decreased with increasing dilution rate. The regulatory significance of these enzymes in lipid accumulation by C. curvata is discussed.     

Effect of Dyebath Additives on Decolorization Efficiency in the Ozonation of Dyehouse Effluent

It is necessary to study the effect of dyebath additives on decolorization efficiency in order to optimize ozone-based decolorization processes as the consumption of ozone can be reduced through selecting ozone favorable additives. The effect of 5 dyebath additives viz. electrolytes (sodium chloride and sodium sulfate), chelating agent (ethylene diamine tetra acetic acid or EDTA), reducing agent (sodium dithionite), optical brightener (Uvitex BHT), and dispersing agent (Zetex DNVL) was investigated. All of the additives showed synergistic effect as addition of sodium chloride, sodium dithionite and Zetex DN-VL markedly improved decolorization efficiency, but EDTA and optical brightener showed negative effect. Sodium sulfate did not show any positive or negative effect on decolorization efficiency.     

Ellipsoidal neighbourhood outlier factor for distributed anomaly detection in resource constrained networks

Anomaly detection in resource constrained wireless networks is an important challenge for tasks such as intrusion detection, quality assurance and event monitoring applications. The challenge is to detect these interesting events or anomalies in a timely manner, while minimising energy consumption in the network. We propose a distributed anomaly detection architecture, which uses multiple hyperellipsoidal clusters to model the data at each sensor node, and identify global and local anomalies in the network. In particular, a novel anomaly scoring method is proposed to provide a score for each hyperellipsoidal model, based on how remote the ellipsoid is relative to their neighbours. We demonstrate using several synthetic and real datasets that our proposed scheme achieves a higher detection performance with a significant reduction in communication overhead in the network compared to centralised and existing schemes.     

Viscosity and Thermal Expansion of Mixed-Alkali Borate Glasses

Transformation-range viscosity and thermal expansion measurements were made for five series (Li-Na, Li-K, Li-Cs, Na-Cs, and K-Rb) of mixed-alkali borate glasses containing 30 mol% total alkali oxide. In each case, negative deviations from additivity were observed in the isokom and glass transformation temperatures. Positive deviations from additivity were observed in the thermal expansion coefficients for the Li-Na, Li-K, and K-Rb glasses, while the Li-Cs and Na-Cs glasses exhibited negative deviations. Maximization of the deviation from additivity in the viscosity and glass transformation temperatures occurred when the radius ratio of the two alkali ions was approximately 1.7 to 1.8. Combination of the results of this study with those of earlier studies indicates that the deviations from additivity observed for these properties are independent of the identity of the glass former.     

Lightweight dynamic symmetry breaking

Symmetries in constraint problems present an opportunity for reducing search. This paper presents Lightweight Dynamic Symmetry Breaking, an automatic symmetry breaking method that is efficient enough to be used as a default, since it never yields a major slowdown while often giving major performance improvements. This is achieved by automatically exploiting certain kinds of symmetry that are common, can be compactly represented, easily and efficiently processed, automatically detected, and lead to large reductions in search. Moreover, the method is easy to implement and integrate in any constraint system. Experimental results show the method is competitive with the best symmetry breaking methods without risking poor performance.     

Exploring aspects of cell intelligence with artificial reaction networks

The Artificial Reaction Network (ARN) is a Cell Signalling Network inspired connectionist representation belonging to the branch of A-Life known as Artificial Chemistry. Its purpose is to represent chemical circuitry and to explore computational properties responsible for generating emergent high-level behaviour associated with cells. In this paper, the computational mechanisms involved in pattern recognition and spatio-temporal pattern generation are examined in robotic control tasks. The results show that the ARN has application in limbed robotic control and computational functionality in common with Artificial Neural Networks. Like spiking neural models, the ARN can combine pattern recognition and complex temporal control functionality in a single network, however it offers increased flexibility. Furthermore, the results illustrate parallels between emergent neural and cell intelligence.     

Adaptive relationships of epoxide hydrolase in herbivorous arthropods

Epoxide hydrolase catalyzes a simple hydrolysis of reactive cyclic ethers that may otherwise alkylate and impair critical proteins and nucleic acids required for life. Although much less studied than the cytochrome P-450 monooxygenases that produce epoxides, differences in subcellular, tissue, pH, substrate, and inhibitor specificities argue for at least three forms of insect epoxide hydrolase. Increasing numbers of epoxides are being identified as plant allelochemicals, antifeedants, and essential hormones or precursors for herbivorous arthropods, and in many cases an associated alkene to diol pathway of metabolism is found. A role for epoxide hydrolase in arthropod-plant interactions is strongly supported by species comparisons and by age-activity and induction studies. Two major limitations for study in biochemical ecology of epoxide hydrolase are the lack of an effective in vivo inhibitor and a range of commercially available radiolabeled substrates for the enzymes.