Elastic net orthogonal forward regression

An efficient two-level model identification method aiming at maximising a model׳s generalisation capability is proposed for a large class of linear-in-the-parameters models from the observational data. A new elastic net orthogonal forward regression (ENOFR) algorithm is employed at the lower level to carry out simultaneous model selection and elastic net parameter estimation. The two regularisation parameters in the elastic net are optimised using a particle swarm optimisation (PSO) algorithm at the upper level by minimising the leave one out (LOO) mean square error (LOOMSE). There are two elements of original contributions. Firstly an elastic net cost function is defined and applied based on orthogonal decomposition, which facilitates the automatic model structure selection process with no need of using a predetermined error tolerance to terminate the forward selection process. Secondly it is shown that the LOOMSE based on the resultant ENOFR models can be analytically computed without actually splitting the data set, and the associate computation cost is small due to the ENOFR procedure. Consequently a fully automated procedure is achieved without resort to any other validation data set for iterative model evaluation. Illustrative examples are included to demonstrate the effectiveness of the new approaches.

Effective interventions: 25 years outside the mainstream

25 years ago when the Durham conferences were in full swing, I presented results of investigations on language and behaviour in autism. I tentatively proposed that early language in autism might tell us about the cognitive skills of people with ASD and the behaviour might lead to greater understanding of which brain systems might be affected. In this presentation, I will update these topics and present a summary of other work I have been involved with in attempting to improve the lives of people with autism and their families. Data on three people with autism at the early stages of speech development showed an unusual pattern of learning colour and number names early. One possibility was that this skill represented a sign of weak central coherence – they only attended to one dimension. Colleagues of mine were equally puzzled so we tried to find out if my results could be replicated – they were not (see Schafer, Williams & Smith, 2014). Instead we found this pattern was also seen in Down Syndrome, but that early vocabulary in autism was associated with low Colorado Meaningfulness at least in comprehension. The Colorado Meaningfulness of a word is a measure of how many words can be associated with it and often involve extensive use of context. Our data suggest that the number of contexts in which a particular word can appear has a role in determining vocabulary in ASD which is consistent with the weak central coherence theory of autism. In the course of this work I also came across a group of young people with autism who appeared to have a written vocabulary but not a spoken one. It seems possible that print might be a medium of communication when speech is not. Repetitive behaviour in autism remains a mystery. We can use functional analysis to determine why the behaviour occurs, but a worryingly large percentage of behaviours are described as being internally driven or sensory reinforced. What does that mean in terms of brain activity – could it be system analogous to epilepsy, where brain activity becomes inappropriately synchronised? At the moment I cannot claim to have solved this problem, but if sensation is a driver then sensory interventions should make a difference. Data from a recent study will be presented to suggest that for some individuals this is the case. Social behaviour remains the key however, and it remains to be seen whether it is possible for social behaviour to be aided. One route that has potential is direct teaching of skills through drama and working with others who do not have social difficulties of the same type. The picture is complicated by changes in social skills with age and experience, but the failure of people with ASD to interact when in settings of social contact is little researched.

Lattice Boltzmann simulation of viscous fluid systems with elastic boundaries

A lattice Boltzmann model able to simulate viscous fluid systems with elastic and movable boundaries is proposed. By introducing the virtual distribution function at the boundary, the Galilean invariance is recovered for the full system. As examples of application, the how in elastic vessels is simulated with the pressure-radius relationship similar to that of the pulmonary blood vessels. The numerical results for steady how are in good agreement with the analytical prediction, while the simulation results for pulsative how agree with the experimental observation of the aortic flows qualitatively. The approach has potential application in the study of the complex fluid systems such as the suspension system as well as the arterial blood flow.

Effective separation of Am(III) and Eu(III) from HNO3 Q1 Q2 solutions using CyMe4-BTPhen-functionalized silica-coated magnetic nanoparticles

It has been shown that CyMe4-BTPhen-functionalized silica-coated maghemite (c-Fe2O3) magnetic nanoparticles (MNPs) are capable of quantitative separation of Am(III) from Eu(III) from HNO3 solutions. These MNPs also show a small but significant selectivity for Am(III) over Cm(III) with a separation factor of around 2 in 4 M HNO3. The water molecule in the cavity of the BTPhen may also play an important part in the selectivity.

Bis amide-aromatic-ureasd - highly effective hydro- and organogelator systems

A series of hydro- and organo-supergelators have been synthesised via coupling of simple bis aromaticureas via alkyl amide linkages. These bis amide-aromatic-ureas exhibited reduced critical gelator concentrations, improved gelator stability, mechanical and dye removal properties for potential use in water purification, in comparison to related bis aromatic-ureas. Systematic structure studies via variation of the bis amide-aromatic-urea linker length as well as functionalization of the terminal aromatic moieties have enabled control over the gel properties.

Bis amide-aromatic-ureasd highly effective hydro- and organogelator systems

A series of hydro- and organo-supergelators have been synthesised via coupling of simple bis aromaticureas via alkyl amide linkages. These bis amide-aromatic-ureas exhibited reduced critical gelator concentrations, improved gelator stability, mechanical and dye removal properties for potential use in water purification, in comparison to related bis aromatic-ureas. Systematic structure studies via variation of the bis amide-aromatic-urea linker length as well as functionalization of the terminal aromatic moieties have enabled control over the gel properties.

Effective separation of Am(III) and Eu(III) from HNO3 solutions using CyMe4-BTPhen-functionalized silica-coated magnetic nanoparticles

It has been shown that CyMe4-BTPhen-functionalized silica-coated maghemite (c-Fe2O3) magnetic nanoparticles (MNPs) are capable of quantitative separation of Am(III) from Eu(III) from HNO3 solutions. These MNPs also show a small but significant selectivity for Am(III) over Cm(III) with a separation factor of around 2 in 4 M HNO3. The water molecule in the cavity of the BTPhen may also play an important part in the selectivity.

Effective separation of the actinides Am(III) and Cm(III) by electronic modulation of bis-(1,2,4-triazin-3-yl)phenanthrolines

It has been shown that modification of the phenanthroline backbone of CyMe4-BTPhen leads to subtle electronic modulation, permitting differential ligation of Am(III) and Cm(III) resulting in separation factors up to 7.

Virtualization for cost-effective teaching of assembly language programming

A virtual system that emulates an ARM-based processor machine has been created to replace a traditional hardware-based system for teaching assembly language. The proposed virtual system integrates, in a single environment, all the development tools necessary to deliver introductory or advanced courses on modern assembly language programming. The virtual system runs a Linux operating system in either a graphical or console mode on a Windows or Linux host machine. No software licenses or extra hardware are required to use the virtual system, thus students are free to carry their own ARM emulator with them on a USB memory stick. Institutions adopting this, or a similar virtual system, can also benefit by reducing capital investment in hardware-based development kits and enable distance learning courses.

Influence of inherent structure shear stress of supercooled liquids on their shear moduli

Configurations of supercooled liquids residing in their local potential minimum (i.e. in their inherent structure, IS) were found to support a non-zero shear stress. This IS stress was attributed to the constraint to the energy minimization imposed by boundary conditions, which keep size and shape of the simulation cell fixed. In this paper we further investigate the influence of these boundary conditions on the IS stress. We investigate its importance for the computation of the low frequency shear modulus of a glass obtaining a consistent picture for the low- and high frequency shear moduli over the full temperature range. Hence, we find that the IS stress corresponds to a non-thermal contribution to the fluctuation term in the Born-Green expression. This leads to an unphysical divergence of the moduli in the low temperature limit if no proper correction for this term is applied. Furthermore, we clarify the IS stress dependence on the system size and put its origin on a more formal basis.

Mesomorphic behaviour in copoly(ester-imide)s of poly(butylene-2,6-naphthalate) (PBN)

Copolycondensation of N,N’-bis(4-hydroxybutyl)-biphenyl-3,4,3',4'-tetracarboxylic diimide at 20 and 25 mol% with bis(4-hydroxybutyl)-2,6-naphthalate produces PBN-based copoly(ester-imide)s that not only crystallise but also form a (smectic) mesophase upon cooling from the melt. Incorporation of 25 mol% imide in PBN causes the glass transition temperature (measured by DSC) to rise from 51 to 74 °C, a significant increase relative to PBN. Furthermore, increased storage- (G'), loss- (G'') and elastic (E) moduli are observed for both copoly(ester-imide)s when compared to PBN itself. Structural analysis of the 20 mol% copolymer by X-ray powder and fibre diffraction, interfaced to computational modelling, suggests a crystal structure related to that of α-PBN, in space group P-1, with cell dimensions a = 4.74, b = 6.38, c = 14.45 Å, α = 106.1, β = 122.1, γ = 97.3°, ρ = 1.37 g cm-3.

Landscape connectivity: a key to effective habitat restoration in lowland agricultural landscapes

Landscape scale habitat restoration has the potential to reconnect habitats in fragmented landscapes. This study investigates landscape connectivity as a key to effective habitat restoration in lowland agricultural landscapes and applies these findings to transferable management recommendations. The study area is the Stonehenge World Heritage Site, UK, where landscape scale chalk grassland restoration has been implemented. Here, the ecological benefits of landscape restoration and the species, habitat and landscape characteristics that facilitate or impede the enhancement of biodiversity and landscape connectivity were investigated. Lepidoptera were used as indictors of restoration success and results showed restoration grasslands approaching the ecological conditions of the target chalk grassland habitat and increasing in biodiversity values within a decade. Restoration success is apparent for four species with a broad range of grass larval host plants (e.g. Melanargia galathea, Maniola jurtina) or with intermediate mobility (Polyommatus icarus). However, two species with specialist larval host plants and low mobility (Lysandra bellargus), are restricted to chalk grassland fragments. Studies of restoration grassland of different ages show that recent grassland restoration (1 or 2 years old) may reduce the functional isolation of chalk grassland fragments. A management experiment showed that mowing increases boundary following behaviour in two species of grassland Lepidoptera; Maniola jurtina and Zygaena filipendulae. Analysis of the landscape scale implications of the grassland restoration illustrates an increase in grassland habitat network size and in landscape connectivity, which is likely to benefit the majority of grassland associated Lepidoptera. Landscape and habitat variables can be managed to increase the success of restoration projects including the spatial targeting of receptor sites, vegetation structure and selection of seed source and management recommendations are provided that are transferrable to other species-rich grassland landscape scale restoration projects. Overall results show restoration success for some habitats and species within a decade. However, additional management is required to assist the re-colonisation of specialist species. Despite this, habitat restoration at the landscape scale can be an effective, long term approach to enhance butterfly biodiversity and landscape connectivity.

Hydrophilic sulfonated bis-1,2,4-triazine ligands are highly effective reagents for separating actinides(iii) from lanthanides(iii) via selective formation of aqueous actinide complexes

We report the first examples of hydrophilic 6,6′-bis(1,2,4-triazin-3-yl)-2,2′-bipyridine (BTBP) and 2,9-bis(1,2,4-triazin-3-yl)-1,10-phenanthroline (BTPhen) ligands, and their applications as actinide(III) selective aqueous complexing agents. The combination of a hydrophobic diamide ligand in the organic phase and a hydrophilic tetrasulfonated bis-triazine ligand in the aqueous phase is able to separate Am(III) from Eu(III) by selective Am(III) complex formation across a range of nitric acid concentrations with very high selectivities, and without the use of buffers. In contrast, disulfonated bis-triazine ligands are unable to separate Am(III) from Eu(III) in this system. The greater ability of the tetrasulfonated ligands to retain Am(III) selectively in the aqueous phase than the corresponding disulfonated ligands appears to be due to the higher aqueous solubilities of the complexes of the tetrasulfonated ligands with Am(III). The selectivities for Am(III) complexation observed with hydrophilic tetrasulfonated bis-triazine ligands are in many cases far higher than those found with the polyaminocarboxylate ligands previously used as actinide-selective complexing agents, and are comparable to those found with the parent hydrophobic bis-triazine ligands. Thus we demonstrate a feasible alternative method to separate actinides from lanthanides than the widely studied approach of selective actinide extraction with hydrophobic bis-1,2,4-triazine ligands such as CyMe4-BTBP and CyMe4-BTPhen.