Learning Fine Motion by Markov Mixtures of Experts

Compliant control is a standard method for performing fine manipulation tasks, like grasping and assembly, but it requires estimation of the state of contact between the robot arm and the objects involved. Here we present a method to learn a model of the movement from measured data. The method requires little or no prior knowledge and the resulting model explicitly estimates the state of contact. The current state of contact is viewed as the hidden state variable of a discrete HMM. The control dependent transition probabilities between states are modeled as parametrized functions of the measurement We show that their parameters can be estimated from measurements concurrently with the estimation of the parameters of the movement in each state of contact. The learning algorithm is a variant of the EM procedure. The E step is computed exactly; solving the M step exactly would require solving a set of coupled nonlinear algebraic equations in the parameters. Instead, gradient ascent is used to produce an increase in likelihood.

Prediction of soil organic partition coefficients by a soil leaching column chromatographic method

The soil organic partition coefficient (K-oc) is one of the most important parameters to depict the transfer and fate of a chemical in the soil-water system. Predicting K-oc by using a chromatographic technique has been developing into a convenient and low-cost method. In this paper, a soil leaching column chromatograpy (SLCC) method employing the soil column packed with reference soil GSE 17201 (obtained from Bayer Landwirtschaftszentrum, Monheim, Germany) and methanol-water eluents was developed to predict the K-oc of hydrophobic organic chemicals (HOCs), over a log K-oc range of 4.8 orders of magnitude, from their capacity factors. The capacity factor with water as an eluent (k(w)') could be obtained by linearly extrapolating capacity factors in methanol-water eluents (k') with various volume fractions of methanol (phi). The important effects of solute activity coefficients in water on k(w)' and K-oc were illustrated. Hence, the correlation between log K-oc and log k(w)' (and log k') exists in the soil. The correlation coefficient (r) of the log K-oc vs. log k(w)' correlation for 58 apolar and polar compounds could reach 0.987, while the correlation coefficients of the log K-oc-log k' correlations were no less than 0.968, with phi ranging from 0 to 0.50. The smaller the phi, the higher the r. Therefore, it is recommended that the eluent of smaller phi, such as water, be used for accurately estimating K-oc. Correspondingly, the r value of the log K-oc-log k(w)' correlation on a reversed-phase Hypersil ODS (Thermo Hypersil, Kleinostheim, Germany) column was less than 0.940 for the same solutes. The SLCC method could provide a more reliable route to predict K-oc indirectly from a correlation with k(w)' than the reversed-phase liquid chromatographic (RPLC) one.

Fine-Grained Layered Multicast

Traditional approaches to receiver-driven layered multicast have advocated the benefits of cumulative layering, which can enable coarse-grained congestion control that complies with TCP-friendliness equations over large time scales. In this paper, we quantify the costs and benefits of using non-cumulative layering and present a new, scalable multicast congestion control scheme which provides a fine-grained approximation to the behavior of TCP additive increase/multiplicative decrease (AIMD). In contrast to the conventional wisdom, we demonstrate that fine-grained rate adjustment can be achieved with only modest increases in the number of layers and aggregate bandwidth consumption, while using only a small constant number of control messages to perform either additive increase or multiplicative decrease.

Fuzzy ART

Adaptive Resonance Theory (ART) models are real-time neural networks for category learning, pattern recognition, and prediction. Unsupervised fuzzy ART and supervised fuzzy ARTMAP synthesize fuzzy logic and ART networks by exploiting the formal similarity between the computations of fuzzy subsethood and the dynamics of ART category choice, search, and learning. Fuzzy ART self-organizes stable recognition categories in response to arbitrary sequences of analog or binary input patterns. It generalizes the binary ART 1 model, replacing the set-theoretic: intersection (∩) with the fuzzy intersection (∧), or component-wise minimum. A normalization procedure called complement coding leads to a symmetric: theory in which the fuzzy inter:>ec:tion and the fuzzy union (∨), or component-wise maximum, play complementary roles. Complement coding preserves individual feature amplitudes while normalizing the input vector, and prevents a potential category proliferation problem. Adaptive weights :otart equal to one and can only decrease in time. A geometric interpretation of fuzzy AHT represents each category as a box that increases in size as weights decrease. A matching criterion controls search, determining how close an input and a learned representation must be for a category to accept the input as a new exemplar. A vigilance parameter (p) sets the matching criterion and determines how finely or coarsely an ART system will partition inputs. High vigilance creates fine categories, represented by small boxes. Learning stops when boxes cover the input space. With fast learning, fixed vigilance, and an arbitrary input set, learning stabilizes after just one presentation of each input. A fast-commit slow-recode option allows rapid learning of rare events yet buffers memories against recoding by noisy inputs. Fuzzy ARTMAP unites two fuzzy ART networks to solve supervised learning and prediction problems. A Minimax Learning Rule controls ARTMAP category structure, conjointly minimizing predictive error and maximizing code compression. Low vigilance maximizes compression but may therefore cause very different inputs to make the same prediction. When this coarse grouping strategy causes a predictive error, an internal match tracking control process increases vigilance just enough to correct the error. ARTMAP automatically constructs a minimal number of recognition categories, or "hidden units," to meet accuracy criteria. An ARTMAP voting strategy improves prediction by training the system several times using different orderings of the input set. Voting assigns confidence estimates to competing predictions given small, noisy, or incomplete training sets. ARPA benchmark simulations illustrate fuzzy ARTMAP dynamics. The chapter also compares fuzzy ARTMAP to Salzberg's Nested Generalized Exemplar (NGE) and to Simpson's Fuzzy Min-Max Classifier (FMMC); and concludes with a summary of ART and ARTMAP applications.

Space, Time and Learning in the Hippocampus: How Fine Spatial and Temporal Scales Are Expanded into Population Codes for Behavioral Control

The hippocampus participates in multiple functions, including spatial navigation, adaptive timing, and declarative (notably, episodic) memory. How does it carry out these particular functions? The present article proposes that hippocampal spatial and temporal processing are carried out by parallel circuits within entorhinal cortex, dentate gyrus, and CA3 that are variations of the same circuit design. In particular, interactions between these brain regions transform fine spatial and temporal scales into population codes that are capable of representing the much larger spatial and temporal scales that are needed to control adaptive behaviors. Previous models of adaptively timed learning propose how a spectrum of cells tuned to brief but different delays are combined and modulated by learning to create a population code for controlling goal-oriented behaviors that span hundreds of milliseconds or even seconds. Here it is proposed how projections from entorhinal grid cells can undergo a similar learning process to create hippocampal place cells that can cover a space of many meters that are needed to control navigational behaviors. The suggested homology between spatial and temporal processing may clarify how spatial and temporal information may be integrated into an episodic memory.

Miniature Eye Movements Enhance Fine Spatial Details

Our eyes are constantly in motion. Even during visual fixation, small eye movements continually jitter the location of gaze. It is known that visual percepts tend to fade when retinal image motion is eliminated in the laboratory. However, it has long been debated whether, during natural viewing, fixational eye movements have functions in addition to preventing the visual scene from fading. In this study, we analysed the influence in humans of fixational eye movements on the discrimination of gratings masked by noise that has a power spectrum similar to that of natural images. Using a new method of retinal image stabilization18, we selectively eliminated the motion of the retinal image that normally occurs during the intersaccadic intervals of visual fixation. Here we show that fixational eye movements improve discrimination of high spatial frequency stimuli, but not of low spatial frequency stimuli. This improvement originates from the temporal modulations introduced by fixational eye movements in the visual input to the retina, which emphasize the high spatial frequency harmonics of the stimulus. In a natural visual world dominated by low spatial frequencies, fixational eye movements appear to constitute an effective sampling strategy by which the visual system enhances the processing of spatial detail.

An investigation on the use of SNR distributions for the optimisation of coarse-fine spectrum sensing for cognitive radio

This thesis investigates the optimisation of Coarse-Fine (CF) spectrum sensing architectures under a distribution of SNRs for Dynamic Spectrum Access (DSA). Three different detector architectures are investigated: the Coarse-Sorting Fine Detector (CSFD), the Coarse-Deciding Fine Detector (CDFD) and the Hybrid Coarse-Fine Detector (HCFD). To date, the majority of the work on coarse-fine spectrum sensing for cognitive radio has focused on a single value for the SNR. This approach overlooks the key advantage that CF sensing has to offer, namely that high powered signals can be easily detected without extra signal processing. By considering a range of SNR values, the detector can be optimised more effectively and greater performance gains realised. This work considers the optimisation of CF spectrum sensing schemes where the security and performance are treated separately. Instead of optimising system performance at a single, constant, low SNR value, the system instead is optimised for the average operating conditions. The security is still provided such that at the low SNR values the safety specifications are met. By decoupling the security and performance, the system’s average performance increases whilst maintaining the protection of licensed users from harmful interference. The different architectures considered in this thesis are investigated in theory, simulation and physical implementation to provide a complete overview of the performance of each system. This thesis provides a method for estimating SNR distributions which is quick, accurate and relatively low cost. The CSFD is modelled and the characteristic equations are found for the CDFD scheme. The HCFD is introduced and optimisation schemes for all three architectures are proposed. Finally, using the Implementing Radio In Software (IRIS) test-bed to confirm simulation results, CF spectrum sensing is shown to be significantly quicker than naive methods, whilst still meeting the required interference probability rates and not requiring substantial receiver complexity increases.

Differentiation and fine structure of larval epaulets in the echinoid Paracentrotus lividus

info:eu-repo/semantics/published

Surface area, porosity and water adsorption properties of fine volcanic ash particles

Our understanding on how ash particles in volcanic plumes react with coexisting gases and aerosols is still rudimentary, despite the importance of these reactions in influencing the chemistry and dynamics of a plume. In this study, six samples of fine ash (<100 μm) from different volcanoes were measured for their specific surface area, as, porosity and water adsorption properties with the aim to provide insights into the capacity of silicate ash particles to react with gases, including water vapour. To do so, we performed high-resolution nitrogen and water vapour adsorption/desorption experiments at 77 K and 303 K, respectively. The nitrogen data indicated as values in the range 1.1-2.1 m2/g, except in one case where as of 10 m2/g was measured. This high value is attributed to incorporation of hydrothermal phases, such as clay minerals, in the ash surface composition. The data also revealed that the ash samples are essentially non-porous, or have a porosity dominated by macropores with widths >500 Å All the specimens had similar pore size distributions, with a small peak centered around 50 Å These findings suggest that fine ash particles have relatively undifferentiated surface textures, irrespective of the chemical composition and eruption type. Adsorption isotherms for water vapour revealed that the capacity of the ash samples for water adsorption is systematically larger than predicted from the nitrogen adsorption as values. Enhanced reactivity of the ash surface towards water may result from (i) hydration of bulk ash constituents; (ii) hydration of surface compounds; and/or (iii) hydroxylation of the surface of the ash. The later mechanism may lead to irreversible retention of water. Based on these experiments, we predict that volcanic ash is covered by a complete monolayer of water under ambient atmospheric conditions. In addition, capillary condensation within ash pores should allow for deposition of condensed water on to ash particles before water reaches saturation in the plume. The total mass of water vapour retained by 1 g of fine ash at 0.95 relative water vapour pressure is calculated to be ∼10-2 g. Some volcanic implications of this study are discussed. © Springer-Verlag 2004.

Automatic and effective multi-dimensional parallelisation of structured mesh based codes

The most common parallelisation strategy for many Computational Mechanics (CM) (typified by Computational Fluid Dynamics (CFD) applications) which use structured meshes, involves a 1D partition based upon slabs of cells. However, many CFD codes employ pipeline operations in their solution procedure. For parallelised versions of such codes to scale well they must employ two (or more) dimensional partitions. This paper describes an algorithmic approach to the multi-dimensional mesh partitioning in code parallelisation, its implementation in a toolkit for almost automatically transforming scalar codes to parallel form, and its testing on a range of ‘real-world’ FORTRAN codes. The concept of multi-dimensional partitioning is straightforward, but non-trivial to represent as a sufficiently generic algorithm so that it can be embedded in a code transformation tool. The results of the tests on fine real-world codes demonstrate clear improvements in parallel performance and scalability (over a 1D partition). This is matched by a huge reduction in the time required to develop the parallel versions when hand coded – from weeks/months down to hours/days.