Fast Learning VIEWNET Architectures for Recognizing 3-D Objects from Multiple 2-D Views

The recognition of 3-D objects from sequences of their 2-D views is modeled by a family of self-organizing neural architectures, called VIEWNET, that use View Information Encoded With NETworks. VIEWNET incorporates a preprocessor that generates a compressed but 2-D invariant representation of an image, a supervised incremental learning system that classifies the preprocessed representations into 2-D view categories whose outputs arc combined into 3-D invariant object categories, and a working memory that makes a 3-D object prediction by accumulating evidence from 3-D object category nodes as multiple 2-D views are experienced. The simplest VIEWNET achieves high recognition scores without the need to explicitly code the temporal order of 2-D views in working memory. Working memories are also discussed that save memory resources by implicitly coding temporal order in terms of the relative activity of 2-D view category nodes, rather than as explicit 2-D view transitions. Variants of the VIEWNET architecture may also be used for scene understanding by using a preprocessor and classifier that can determine both What objects are in a scene and Where they are located. The present VIEWNET preprocessor includes the CORT-X 2 filter, which discounts the illuminant, regularizes and completes figural boundaries, and suppresses image noise. This boundary segmentation is rendered invariant under 2-D translation, rotation, and dilation by use of a log-polar transform. The invariant spectra undergo Gaussian coarse coding to further reduce noise and 3-D foreshortening effects, and to increase generalization. These compressed codes are input into the classifier, a supervised learning system based on the fuzzy ARTMAP algorithm. Fuzzy ARTMAP learns 2-D view categories that are invariant under 2-D image translation, rotation, and dilation as well as 3-D image transformations that do not cause a predictive error. Evidence from sequence of 2-D view categories converges at 3-D object nodes that generate a response invariant under changes of 2-D view. These 3-D object nodes input to a working memory that accumulates evidence over time to improve object recognition. ln the simplest working memory, each occurrence (nonoccurrence) of a 2-D view category increases (decreases) the corresponding node's activity in working memory. The maximally active node is used to predict the 3-D object. Recognition is studied with noisy and clean image using slow and fast learning. Slow learning at the fuzzy ARTMAP map field is adapted to learn the conditional probability of the 3-D object given the selected 2-D view category. VIEWNET is demonstrated on an MIT Lincoln Laboratory database of l28x128 2-D views of aircraft with and without additive noise. A recognition rate of up to 90% is achieved with one 2-D view and of up to 98.5% correct with three 2-D views. The properties of 2-D view and 3-D object category nodes are compared with those of cells in monkey inferotemporal cortex.

Fuzzy ART: Fast Stable Learning and Categorization of Analog Patterns by an Adaptive Resonance System

A Fuzzy ART model capable of rapid stable learning of recognition categories in response to arbitrary sequences of analog or binary input patterns is described. Fuzzy ART incorporates computations from fuzzy set theory into the ART 1 neural network, which learns to categorize only binary input patterns. The generalization to learning both analog and binary input patterns is achieved by replacing appearances of the intersection operator (n) in AHT 1 by the MIN operator (Λ) of fuzzy set theory. The MIN operator reduces to the intersection operator in the binary case. Category proliferation is prevented by normalizing input vectors at a preprocessing stage. A normalization procedure called complement coding leads to a symmetric theory in which the MIN operator (Λ) and the MAX operator (v) of fuzzy set theory play complementary roles. Complement coding uses on-cells and off-cells to represent the input pattern, and preserves individual feature amplitudes while normalizing the total on-cell/off-cell vector. Learning is stable because all adaptive weights can only decrease in time. Decreasing weights correspond to increasing sizes of category "boxes". Smaller vigilance values lead to larger category boxes. Learning stops when the input space is covered by boxes. With fast learning and a finite input set of arbitrary size and composition, learning stabilizes after just one presentation of each input pattern. A fast-commit slow-recode option combines fast learning with a forgetting rule that buffers system memory against noise. Using this option, rare events can be rapidly learned, yet previously learned memories are not rapidly erased in response to statistically unreliable input fluctuations.

Adaptive Resonance Associative Map: A Hierarchical ART System for Fast Stable Associative Learning

This paper introduces a new class of predictive ART architectures, called Adaptive Resonance Associative Map (ARAM) which performs rapid, yet stable heteroassociative learning in real time environment. ARAM can be visualized as two ART modules sharing a single recognition code layer. The unit for recruiting a recognition code is a pattern pair. Code stabilization is ensured by restricting coding to states where resonances are reached in both modules. Simulation results have shown that ARAM is capable of self-stabilizing association of arbitrary pattern pairs of arbitrary complexity appearing in arbitrary sequence by fast learning in real time environment. Due to the symmetrical network structure, associative recall can be performed in both directions.

Rapid separation of biologically important low molecular weight compounds by microchip electrophoresis and ultra-fast performance liquid chromatography

The research work in this thesis reports rapid separation of biologically important low molecular weight compounds by microchip electrophoresis and ultrahigh liquid chromatography. Chapter 1 introduces the theory and principles behind capillary electrophoresis separation. An overview of the history, different modes and detection techniques coupled to CE is provided. The advantages of microchip electrophoresis are highlighted. Some aspects of metal complex analysis by capillary electrophoresis are described. Finally, the theory and different modes of the liquid chromatography technology are presented. Chapter 2 outlines the development of a method for the capillary electrophoresis of (R, S) Naproxen. Variable parameters of the separation were optimized (i.e. buffer concentration and pH, concentration of chiral selector additives, applied voltage and injection condition).The method was validated in terms of linearity, precision, and LOD. The optimized method was then transferred to a microchip electrophoresis system. Two different types of injection i.e. gated and pinched, were investigated. This microchip method represents the fastest reported chiral separation of Naproxen to date. Chapter 3 reports ultra-fast separation of aromatic amino acid by capillary electrophoresis using the short-end technique. Variable parameters of the separation were optimized and validated. The optimized method was then transferred to a microchip electrophoresis system where the separation time was further reduced. Chapter 4 outlines the use of microchip electrophoresis as an efficient tool for analysis of aluminium complexes. A 2.5 cm channel with linear imaging UV detection was used to separate and detect aluminium-dopamine complex and free dopamine. For the first time, a baseline, separation of aluminium dopamine was achieved on a 15 seconds timescale. Chapter 5 investigates a rapid, ultra-sensitive and highly efficient method for quantification of histamine in human psoriatic plaques using microdialysis and ultrahigh performance liquid chromatography with fluorescence detection. The method utilized a sub-two-micron packed C18 stationary phase. A fluorescent reagent, 4-(1-pyrene) butyric acid N-hydroxysuccinimide ester was conjugated to the primary and secondary amino moieties of histamine. The dipyrene-labeled histamine in human urine was also investigated by ultrahigh pressure liquid chromatography using a C18 column with 1.8 μm particle diameter. These methods represent one of the fastest reported separations to date of histamine using fluorescence detection.

High-fidelity, broadband stimulated-Brillouin-scattering-based slow light using fast noise modulation.

We demonstrate a 5-GHz-broadband tunable slow-light device based on stimulated Brillouin scattering in a standard highly-nonlinear optical fiber pumped by a noise-current-modulated laser beam. The noisemodulation waveform uses an optimized pseudo-random distribution of the laser drive voltage to obtain an optimal flat-topped gain profile, which minimizes the pulse distortion and maximizes pulse delay for a given pump power. In comparison with a previous slow-modulation method, eye-diagram and signal-to-noise ratio (SNR) analysis show that this broadband slow-light technique significantly increases the fidelity of a delayed data sequence, while maintaining the delay performance. A fractional delay of 0.81 with a SNR of 5.2 is achieved at the pump power of 350 mW using a 2-km-long highly nonlinear fiber with the fast noise-modulation method, demonstrating a 50% increase in eye-opening and a 36% increase in SNR in the comparison.

Measuring fast gene dynamics in single cells with time-lapse luminescence microscopy.

Time-lapse fluorescence microscopy is an important tool for measuring in vivo gene dynamics in single cells. However, fluorescent proteins are limited by slow chromophore maturation times and the cellular autofluorescence or phototoxicity that arises from light excitation. An alternative is luciferase, an enzyme that emits photons and is active upon folding. The photon flux per luciferase is significantly lower than that for fluorescent proteins. Thus time-lapse luminescence microscopy has been successfully used to track gene dynamics only in larger organisms and for slower processes, for which more total photons can be collected in one exposure. Here we tested green, yellow, and red beetle luciferases and optimized substrate conditions for in vivo luminescence. By combining time-lapse luminescence microscopy with a microfluidic device, we tracked the dynamics of cell cycle genes in single yeast with subminute exposure times over many generations. Our method was faster and in cells with much smaller volumes than previous work. Fluorescence of an optimized reporter (Venus) lagged luminescence by 15-20 min, which is consistent with its known rate of chromophore maturation in yeast. Our work demonstrates that luciferases are better than fluorescent proteins at faithfully tracking the underlying gene expression.

Can Genetics Predict Response to Complex Behavioral Interventions? Evidence from a Genetic Analysis of the Fast Track Randomized Control Trial.

Early interventions are a preferred method for addressing behavioral problems in high-risk children, but often have only modest effects. Identifying sources of variation in intervention effects can suggest means to improve efficiency. One potential source of such variation is the genome. We conducted a genetic analysis of the Fast Track randomized control trial, a 10-year-long intervention to prevent high-risk kindergarteners from developing adult externalizing problems including substance abuse and antisocial behavior. We tested whether variants of the glucocorticoid receptor gene NR3C1 were associated with differences in response to the Fast Track intervention. We found that in European-American children, a variant of NR3C1 identified by the single-nucleotide polymorphism rs10482672 was associated with increased risk for externalizing psychopathology in control group children and decreased risk for externalizing psychopathology in intervention group children. Variation in NR3C1 measured in this study was not associated with differential intervention response in African-American children. We discuss implications for efforts to prevent externalizing problems in high-risk children and for public policy in the genomic era.

Big or fast: two strategies in the developmental control of body size.

Adult body size is controlled by the mechanisms that stop growth when a species-characteristic size has been reached. The mechanisms by which size is sensed and by which this information is transduced to the growth regulating system are beginning to be understood in a few species of insects. Two rather different strategies for control have been discovered; one favors large body size and the other favors rapid development.

Fast divide-and-conquer algorithms for preemptive scheduling problems with controllable processing times - a polymatroid optimization approach

We consider a variety of preemptive scheduling problems with controllable processing times on a single machine and on identical/uniform parallel machines, where the objective is to minimize the total compression cost. In this paper, we propose fast divide-and-conquer algorithms for these scheduling problems. Our approach is based on the observation that each scheduling problem we discuss can be formulated as a polymatroid optimization problem. We develop a novel divide-and-conquer technique for the polymatroid optimization problem and then apply it to each scheduling problem. We show that each scheduling problem can be solved in $ \O({\rm T}_{\rm feas}(n) \times\log n)$ time by using our divide-and-conquer technique, where n is the number of jobs and Tfeas(n) denotes the time complexity of the corresponding feasible scheduling problem with n jobs. This approach yields faster algorithms for most of the scheduling problems discussed in this paper.

A fast, streaming SIMD extensions 2, logistic squashing function

Schraudolph proposed an excellent exponential approximation providing increased performance particularly suited to the logistic squashing function used within many neural networking applications. This note applies Intel's streaming SIMD Extensions 2 (SSE2), where SIMD is single instruction multiple data, of the Pentum IV class processor to Schraudolph's technique, further increasing the performance of the logistic squashing function. It was found that the calculation of the new 32-bit SSE2 logistic squashing function described here was up to 38 times faster than the conventional exponential function and up to 16 times faster than a Schraudolph-style 32-bit method on an Intel Pentum D 3.6 GHz CPU.

Are Fast Responses More Random? Testing the Effect of Response Time on Scale in an Online Choice Experiment

Scepticism over stated preference surveys conducted online revolves around the concerns over “professional respondents” who might rush through the questionnaire without sufficiently considering the information provided. To gain insight on the validity of this phenomenon and test the effect of response time on choice randomness, this study makes use of a recently conducted choice experiment survey on ecological and amenity effects of an offshore windfarm in the UK. The positive relationship between self-rated and inferred attribute attendance and response time is taken as evidence for a link between response time and cognitive effort. Subsequently, the generalised multinomial logit model is employed to test the effect of response time on scale, which indicates the weight of the deterministic relative to the error component in the random utility model. Results show that longer response time increases scale, i.e. decreases choice randomness. This positive scale effect of response time is further found to be non-linear and wear off at some point beyond which extreme response time decreases scale. While response time does not systematically affect welfare estimates, higher response time increases the precision of such estimates. These effects persist when self-reported choice certainty is controlled for. Implications of the results for online stated preference surveys and further research are discussed.