This research investigates the paradox of creativity in autism. That is, whether people with subclinical autistic traits have cognitive styles conducive to creativity or whether they are disadvantaged by the implied cognitive and behavioural rigidity of the autism phenotype. The relationship between divergent thinking (a cognitive component of creativity), perception of ambiguous figures, and self-reported autistic traits was evaluated in 312 individuals in a non-clinical sample. High levels of autistic traits were significantly associated with lower fluency scores on the divergent thinking tasks. However autistic traits were associated with high numbers of unusual responses on the divergent thinking tasks. Generation of novel ideas is a prerequisite for creative problem solving and may be an adaptive advantage associated with autistic traits.

In this work, we introduce attention as a state-of-the-art mechanism for classification of radio galaxies, using convolutional neural networks. We present an attention-based model that performs on par with previous classifiers while using more than 50 per cent fewer parameters than the next smallest classic CNN application in this field. We demonstrate quantitatively how the selection of normalization and aggregation methods used in attention-gating can affect the output of individual models, and show that the resulting attention maps can be used to interpret the classification choices made by the model. We observe that the salient regions identified by the our model align well with the regions an expert human classifier would attend to make equivalent classifications. We show that while the selection of normalization and aggregation may only minimally affect the performance of individual models, it can significantly affect the interpretability of the respective attention maps and by selecting a model which aligns well with how astronomers classify radio sources by eye, a user can employ the model in a more effective manner.

This study presents preliminary experimental data suggesting that sodium 4-(pyrene-1-yl)butane-1-sulfonate (PBSA), 5 , an analogue of sodium pyrene-1-sulfonate (PSA), 1 , enhances the stability of aqueous reduced graphene oxide (RGO) graphene dispersions. We find that RGO and exfoliated graphene dispersions prepared in the presence of 5 are approximately double the concentration of those made with commercially available PSA, 1 . Quantum mechanical and molecular dynamics simulations provide key insights into the behavior of these molecules on the graphene surface. The seemingly obvious introduction of a polar sulfonate head group linked via an appropriate alkyl spacer to the aromatic core results in both more efficient binding of 5 to the graphene surface and more efficient solvation of the polar head group by bulk solvent (water). Overall, this improves the stabilization of the graphene flakes by disfavoring dissociation of the stabilizer from the graphene surface and inhibiting reaggregation by electrostatic and steric repulsion. These insights are currently the subject of further investigations in an attempt to develop a rational approach to the design of more effective dispersing agents for rGO and graphene in aqueous solution.

Weight sharing in convolutional neural networks (CNNs) ensures that their feature maps will be translation-equivariant. However, although conventional convolutions are equivariant to translation, they are not equivariant to other isometries of the input image data, such as rotation and reflection. For the classification of astronomical objects such as radio galaxies, which are expected statistically to be globally orientation invariant, this lack of dihedral equivariance means that a conventional CNN must learn explicitly to classify all rotated versions of a particular type of object individually. In this work we present the first application of group-equivariant convolutional neural networks to radio galaxy classification and explore their potential for reducing intra-class variability by preserving equivariance for the Euclidean group E(2), containing translations, rotations and reflections. For the radio galaxy classification problem considered here, we find that classification performance is modestly improved by the use of both cyclic and dihedral models without additional hyper-parameter tuning, and that a D16 equivariant model provides the best test performance. We use the Monte Carlo Dropout method as a Bayesian approximation to recover epistemic uncertainty as a function of image orientation and show that E(2)-equivariant models are able to reduce variations in model confidence as a function of rotation.

No abstract