Dynamic automata for mobile robot learning

Researchers at the University of Reading have developed over many years some simple mobile robots that explore an environment they perceive through simple ultrasonic sensors. Information from these sensors has allowed the robots to learn the simple task of moving around while avoiding dynamic obstacles using a static set of fuzzy automata, the choice of which has been criticised, due to its arbitrary nature. This paper considers how a dynamic set of automata can overcome this criticism. In addition, a new reinforcement learning function is outlined which is both scalable to different numbers and types of sensors. The innovations compare successfully with earlier work.

Evaluation of probabilistic occupancy map people detection for surveillance systems

In this paper, we evaluate the Probabilistic Occupancy Map (POM) pedestrian detection algorithm on the PETS 2009 benchmark dataset. POM is a multi-camera generative detection method, which estimates ground plane occupancy from multiple background subtraction views. Occupancy probabilities are iteratively estimated by fitting a synthetic model of the background subtraction to the binary foreground motion. Furthermore, we test the integration of this algorithm into a larger framework designed for understanding human activities in real environments. We demonstrate accurate detection and localization on the PETS dataset, despite suboptimal calibration and foreground motion segmentation input.

Classification-based probabilistic modeling of texture transition for fast line search tracking and delineation

We introduce a classification-based approach to finding occluding texture boundaries. The classifier is composed of a set of weak learners, which operate on image intensity discriminative features that are defined on small patches and are fast to compute. A database that is designed to simulate digitized occluding contours of textured objects in natural images is used to train the weak learners. The trained classifier score is then used to obtain a probabilistic model for the presence of texture transitions, which can readily be used for line search texture boundary detection in the direction normal to an initial boundary estimate. This method is fast and therefore suitable for real-time and interactive applications. It works as a robust estimator, which requires a ribbon-like search region and can handle complex texture structures without requiring a large number of observations. We demonstrate results both in the context of interactive 2D delineation and of fast 3D tracking and compare its performance with other existing methods for line search boundary detection.

A forward constrained selection algorithm for probabilistic neural network

A new probabilistic neural network (PNN) learning algorithm based on forward constrained selection (PNN-FCS) is proposed. An incremental learning scheme is adopted such that at each step, new neurons, one for each class, are selected from the training samples arid the weights of the neurons are estimated so as to minimize the overall misclassification error rate. In this manner, only the most significant training samples are used as the neurons. It is shown by simulation that the resultant networks of PNN-FCS have good classification performance compared to other types of classifiers, but much smaller model sizes than conventional PNN.

A multi-level probabilistic neural network

Based on the idea of an important cluster, a new multi-level probabilistic neural network (MLPNN) is introduced. The MLPNN uses an incremental constructive approach, i.e. it grows level by level. The construction algorithm of the MLPNN is proposed such that the classification accuracy monotonically increases to ensure that the classification accuracy of the MLPNN is higher than or equal to that of the traditional PNN. Numerical examples are included to demonstrate the effectiveness of proposed new approach.

Roles of the translationally controlled tumour protein (TCTP) and the double-stranded RNA-dependent protein kinase, PKR, in cellular stress responses

Translationally controlled tumour protein (TCTP) is a highly conserved protein present in all eukaryotic organisms. Various cellular functions and molecular interactions have been ascribed to this protein, many related to its growth-promoting and antiapoptotic properties. TCTP levels are highly regulated in response to various cellular stimuli and stresses. We have shown recently that the double-stranded RNA-dependent protein kinase, PKR, is involved in translational regulation of TCTP. Here we extend these studies by demonstrating that TCTP is downregulated in response to various proapoptotic treatments, in particular agents that induce Ca++ stress, in a PKR-dependent manner. This regulation requires phosphorylation of protein synthesis factor eIF2α. Since TCTP has been characterized as an antiapoptotic and Ca++-binding protein, we asked whether it is involved in protecting cells from Ca++-stress-induced apoptosis. Overexpression of TCTP partially protects cells against thapsigargin-induced apoptosis, as measured using caspase-3 activation assays, a nuclear fragmentation assay, using fluorescence-activated cell sorting analysis, and time-lapse video microscopy. TCTP also protects cells against the proapoptotic effects of tunicamycin and etoposide, but not against those of arsenite. Our results imply that cellular TCTP levels influence sensitivity to apoptosis and that PKR may exert its proapoptotic effects at least in part through downregulation of TCTP via eIF2α phosphorylation.

Carnauba wax nanoparticles enhance strong systemic and mucosal cellular and humoral immune responses to HIV-gp140 antigen

Induction of humoral responses to HIV at mucosal compartments without inflammation is important for vaccine design. We developed charged wax nanoparticles that efficiently adsorb protein antigens and are internalized by DC in the absence of inflammation. HIV-gp140-adsorbed nanoparticles induced stronger in vitro T-cell proliferation responses than antigen alone. Such responses were greatly enhanced when antigen was co-adsorbed with TLR ligands. Immunogenicity studies in mice showed that intradermal vaccination with HIV-gp140 antigen-adsorbed nanoparticles induced high levels of specific IgG. Importantly, intranasal immunization with HIV-gp140-adsorbed nanoparticles greatly enhanced serum and vaginal IgG and IgA responses. Our results show that HIV-gp140-carrying wax nanoparticles can induce strong cellular/humoral immune responses without inflammation and may be of potential use as effective mucosal adjuvants for HIV vaccine candidates.

Nuclear Dbf2-related protein kinases (NDRs) in isolated cardiac myocytes and the myocardium: activation by cellular stresses and by phosphoprotein serine-/threonine-phosphatase inhibitors

The nuclear Dbf2-related protein kinases 1 and 2 (NDR1/2) are closely-related AGC family kinases that are strongly conserved through evolution. In mammals, they are activated inter alia by phosphorylation of an hydrophobic domain threonine-residue [NDR1(Thr-444)/NDR2(Thr-442)] by an extrinsic protein kinase followed by autophosphorylation of a catalytic domain serine-residue [NDR1(Ser-281)/NDR2(Ser-282)]. We examined NDR1/2 expression and regulation in primary cultures of neonatal rat cardiac myocytes and in perfused adult rat hearts. In myocytes, transcripts for NDR2, but not NDR1, were induced by the hypertrophic agonist, endothelin-1. NDR1(Thr-444) and NDR2(Thr-442) were rapidly phosphorylated (maximal in 15-30 min) in myocytes exposed to some phosphoprotein Ser-/Thr-phosphatase 1/2 inhibitors (calyculin A, okadaic acid) and, to a lesser extent, by hyperosmotic shock, low concentrations of H(2)O(2), or chelerythrine. In myocytes adenovirally-transduced to express FLAG-NDR2 (which exhibited a mainly-cytoplasmic localisation), the same agents increased FLAG-NDR2 activity as assessed by in vitro protein kinase assays, indicative of FLAG-NDR2(Ser-282/Thr-442) phosphorylation. Calyculin A-induced phosphorylation of NDR1(Thr-444)/NDR2(Thr-442) and activation of FLAG-NDR2 were inhibited by staurosporine, but not by other protein kinase inhibitors tested. In ex vivo rat hearts, NDR1(Thr-444)/NDR2(Thr-442) were phosphorylated in response to ischaemia-reperfusion or calyculin A. From a pathological viewpoint, we conclude that activities of NDR1 and NDR2 are responsive to cytotoxic stresses in heart preparations and this may represent a previously-unidentified response to myocardial ischaemia in vivo.

Beyond antioxidants: the cellular and molecular interactions of flavonoids and how these underpin their actions on the brain

The consumption of flavonoid-rich foods and beverages has been suggested to limit the neurodegeneration associated with a variety of neurological disorders and to prevent or reverse normal or abnormal deteriorations in cognitive performance. Flavonoids mediate these effects via a number of routes, including a potential to protect neurons against injury induced by neurotoxins, an ability to suppress neuroinflammation and a potential to promote memory, learning and cognitive function. Originally, it was thought that such actions were mediated by the antioxidant capacity of flavonoids. However, their limited absorption and their low bioavailability in the brain suggest that this explanation is unlikely. Instead, this multiplicity of effects appears to be underpinned by three separate processes: first, through their interactions with important neuronal and glial signalling cascades in the brain, most notably the phosphatidylinositol 3-kinase/Akt and mitogen-activated protein kinase pathways that regulate pro-survival transcription factors and gene expression; second, through an ability to improve peripheral and cerebral blood flow and to trigger angiogenesis and neurogenesis in the hippocampus; third, by their capacity to directly react with and scavenge neurotoxic species and pro-inflammatory agents produced in the brain as a result of both normal and abnormal brain ageing. The present review explores the potential inhibitory or stimulatory actions of flavonoids within these three systems and describes how such interactions are likely to underlie neurological effects.

Cellular and molecular investigations into the development of the pectoral girdle

The forelimbs of higher vertebrates are composed of two portions: the appendicular region (stylopod, zeugopod and autopod) and the less prominent proximal girdle elements (scapula and clavicle) that brace the limb to the main trunk axis. We show that the formation of the muscles of the proximal limb occurs through two distinct mechanisms. The more superficial girdle muscles (pectoral and latissimus dorsi) develop by the “In–Out” mechanism whereby migration of myogenic cells from the somites into the limb bud is followed by their extension from the proximal limb bud out onto the thorax. In contrast, the deeper girdle muscles (e.g. rhomboideus profundus and serratus anterior) are induced by the forelimb field which promotes myotomal extension directly from the somites. Tbx5 inactivation demonstrated its requirement for the development of all forelimb elements which include the skeletal elements, proximal and distal muscles as well as the sternum in mammals and the cleithrum of fish. Intriguingly, the formation of the diaphragm musculature is also dependent on the Tbx5 programme. These observations challenge our classical views of the boundary between limb and trunk tissues. We suggest that significant structures located in the body should be considered as components of the forelimb.

Path planning of robots in noisy workspaces using learning automata

The problem of a manipulator operating in a noisy workspace and required to move from an initial fixed position P0 to a final position Pf is considered. However, Pf is corrupted by noise, giving rise to Pˆf, which may be obtained by sensors. The use of learning automata is proposed to tackle this problem. An automaton is placed at each joint of the manipulator which moves according to the action chosen by the automaton (forward, backward, stationary) at each instant. The simultaneous reward or penalty of the automata enables avoiding any inverse kinematics computations that would be necessary if the distance of each joint from the final position had to be calculated. Three variable-structure learning algorithms are used, i.e., the discretized linear reward-penalty (DLR-P, the linear reward-penalty (LR-P ) and a nonlinear scheme. Each algorithm is separately tested with two (forward, backward) and three forward, backward, stationary) actions.

Early warning with calibrated and sharper probabilistic forecasts

Given a nonlinear model, a probabilistic forecast may be obtained by Monte Carlo simulations. At a given forecast horizon, Monte Carlo simulations yield sets of discrete forecasts, which can be converted to density forecasts. The resulting density forecasts will inevitably be downgraded by model mis-specification. In order to enhance the quality of the density forecasts, one can mix them with the unconditional density. This paper examines the value of combining conditional density forecasts with the unconditional density. The findings have positive implications for issuing early warnings in different disciplines including economics and meteorology, but UK inflation forecasts are considered as an example.

A 96-well plate fluorescence assay for assessment of cellular permeability and active efflux in Salmonella enterica serovar Typhimurium and Escherichia coli

Multiply antibiotic-resistant (MAR) mutants of Escherichia coli and Salmonella enterica are characterized by reduced susceptibility to several unrelated antibiotics, biocides and other xenobiotics. Porin loss and/or active efflux have been identified as a key mechanisms of MAR. A single rapid test was developed for MAR. The intracellular accumulation of the fluorescent probe Hoechst (H) 33342 (bisbenzimide) by MAR mutants and those with defined disruptions in efflux pump and porin genes was determined in 96-well plate format. The accumulation of H33342 was significantly (P < 0.0001) reduced in MAR mutants of S. enterica serovar Typhimurium (n = 4) and E. coli (n = 3) by 41 +/- 8% and 17.3 +/- 7.2%, respectively, compared with their parental strains, which was reversed by the transmembrane proton gradient-collapsing agent carbonyl cyanide-m-chlorophenyl hydrazone (CCCP) and the efflux pump inhibitor phenylalanine-arginine-beta-naphthylamide (PA beta N). The accumulation of H33342 was significantly reduced in mutants of Salmonella Typhimurium with defined disruptions in genes encoding the porins OmpC, OmpF, OmpX and OmpW, but increased in those with disruptions in efflux pump components TolC, AcrB and AcrF. Reduced accumulation of H33342 in three other MAR mutants of Salmonella Typhimurium correlated with the expression of porin and efflux pump proteins. The intracellular accumulation of H33342 provided a sensitive and specific test for MAR that is cheap and relatively rapid. Differential sensitivity to CCCP and PA beta N provided a further means to phenotypically identify MAR mutants and the role of active efflux in each strain.

Regularized logistic models for probabilistic forecasting and diagnostics

Logistic models are studied as a tool to convert dynamical forecast information (deterministic and ensemble) into probability forecasts. A logistic model is obtained by setting the logarithmic odds ratio equal to a linear combination of the inputs. As with any statistical model, logistic models will suffer from overfitting if the number of inputs is comparable to the number of forecast instances. Computational approaches to avoid overfitting by regularization are discussed, and efficient techniques for model assessment and selection are presented. A logit version of the lasso (originally a linear regression technique), is discussed. In lasso models, less important inputs are identified and the corresponding coefficient is set to zero, providing an efficient and automatic model reduction procedure. For the same reason, lasso models are particularly appealing for diagnostic purposes.