Can buy me love: Mate attraction goals lead to perceptual readiness for status products

Exposure to mating cues activates the goal to signal one's mate value to members of the opposite sex. This mate attraction goal may render men perceptually ready for products that signal their mate value to women. As men's mate value is partly determined by their financial prospects, men may be more likely to notice products that would signal their financial resources to women. The current study demonstrates that exposure to a sexily dressed woman increases single men's likelihood of noticing status products in a visual display. Not only do these findings further support the link between conspicuous consumption and male mating strategies, they are the first to demonstrate perceptual readiness for indirect (i.e., products) rather than direct (i.e., opposite sex individuals) means for reproduction.     

Narrow personality traits and organizational attraction: Evidence for the complementary hypothesis

Although the interactionist perspective has been widely studied in organizational attractiveness, there is no research comparing the explanatory power of the complementary and supplementary hypotheses in predicting attraction. The authors test these perspectives in the context of the instrumental-symbolic framework. The authors also examine whether the use of narrow personality facets, such as Trust (under the Big Five trait Agreeableness), Assertiveness (under Extraversion), and Imagination (under Openness to Experience) enhances the prediction of attraction. Job seekers (N = 220) provided self-ratings of personality, ratings of organizational traits, and their level of attraction to a potential future employer. Results supported predictions based on complementarity, suggesting that organizations adopting a recruiting strategy based on similarity in personality may not succeed in attracting their most preferred candidates. The findings also suggested that narrow facets are useful in predicting attraction, providing further evidence for the predictive benefits of narrow personality traits.     

An eye-tracking attention based model for abstractive text headline

Online network platforms provide great convenience for users to obtain information. However, it’s challenging to select the required information from enormous texts. Automatic text headline generation methods not only guide users to select the information they are interested in, but also solve the problem of information overload. Nevertheless, the existing works mainly utilize the grammar rules to obtain the key information of the source text, while ignoring the dwell time of user’s attention on different text contents. To address this issue, this paper proposes an abstractive text headline generation model based on the eye-tracking attention mechanism. Specifically, this model first relies on the eye-tracking data to establish the mapping relationship between text words and the words’ reading time. Then, an eye-tracking attention mechanism is constructed to judge the importance of different words. Finally, this attention mechanism is integrated into the encoder-decoder framework to generate a high-quality headline. Experimental results obtained from different datasets demonstrate that the headline generated by our model is more concise. Moreover, our proposed model outperforms significantly the classical headline generation models on ROUGE-1, ROUGE-2 and ROUGE-L.     

Multi-tasking deep convolutional network architecture design for extracting nonverbal communicative information from a face

Facial expressions convey not only emotions but also communicative information. Therefore, facial expressions should be analysed to understand communication. The objective of this study is to develop an automatic facial expression analysis system for extracting nonverbal communicative information. This study focuses on specific communicative information: emotions expressed through facial movements and the direction of the expressions. We propose a multi-tasking deep convolutional network (DCN) to classify facial expressions, detect the facial regions, and estimate face angles. We reformulate facial region detection and face angle estimation as regression problems and add task-specific output layers in the DCN’s architecture. Experimental results show that the proposed method performs all tasks accurately. In this study, we show the feasibility of the multi-tasking DCN for extracting nonverbal communicative information from a human face.     

The importance of feature distribution and correlation for simulating 3 to 4-month-old infants' visual categorization processes

Mareschal, French, and Quinn (2000) and Mareschal, Quinn, and French (2002) have proposed a connectionist model of visual categorization in 3- to 4-month-old infants that simulates and predicts previously unexplained behavioural effects such as the asymmetric categorization effect (French, Mareschal, Mermillod, & Quinn, 2004). In the current paper, we show that the model's ability to simulate the asymmetry depends on the correlational structure of the stimuli. These results are important given that adults (Anderson & Fincham, 1996) as well as infants (Younger & Cohen, 1986) are able to rely on correlation information to perform visual categorization. At a behavioural level, the current paper suggests that pure bottom-up processes, based on the correlational structure of the categories, could explain the disappearance of the asymmetry in older 10-month-old infants (Furrer & Younger, 2005). Moreover, our results also raise new challenges for visual categorization models that attempt to simulate the shift from asymmetric categorization in 3- to 4-month-old to symmetric categorization in 10-month-old infants (Shultz & Cohen, 2004; Westermann & Mareschal, 2004, 2012).     

Hierarchies of Self-Organizing Maps for action recognition

We propose a hierarchical neural architecture able to recognise observed human actions. Each layer in the architecture represents increasingly complex human activity features. The first layer consists of a SOM which performs dimensionality reduction and clustering of the feature space. It represents the dynamics of the stream of posture frames in action sequences as activity trajectories over time. The second layer in the hierarchy consists of another SOM which clusters the activity trajectories of the first-layer SOM and learns to represent action prototypes. The third- and last-layer of the hierarchy consists of a neural network that learns to label action prototypes of the second-layer SOM and is independent – to certain extent – of the camera’s angle and relative distance to the actor. The experiments were carried out with encouraging results with action movies taken from the INRIA 4D repository. In terms of representational accuracy, measured as the recognition rate over the training set, the architecture exhibits 100% accuracy indicating that actions with overlapping patterns of activity can be correctly discriminated. On the other hand, the architecture exhibits 53% recognition rate when presented with the same actions interpreted and performed by a different actor. Experiments on actions captured from different view points revealed a robustness of our system to camera rotation. Indeed, recognition accuracy was comparable to the single viewpoint case. To further assess the performance of the system we have also devised a behavioural experiments in which humans were asked to recognise the same set of actions, captured from different points of view. Results form such a behavioural study let us argue that our architecture is a good candidate as cognitive model of human action recognition, as architectural results are comparable to those observed in humans.     

PSO-GA based hybrid with Adam Optimization for ANN training with application in Medical Diagnosis

This paper introduces a novel PSO-GA based hybrid training algorithm with Adam Optimization and contrasts performance with the generic Gradient Descent based Backpropagation algorithm with Adam Optimization for training Artificial Neural Networks. We aim to overcome the shortcomings of the traditional algorithm, such as slower convergence rate and frequent convergence to local minima, by employing the characteristics of evolutionary algorithms. PSO has a property of faster convergence rate, which can be exploited to account for the slower pace of convergence of the traditional BP (which is due to low values of gradients). In contrast, the integration with GA complements the drawback of convergence to local minima as GA, possesses the capability of efficient global search. So by this integration of these algorithms, we propose our new hybrid algorithm for training ANNs. We compare both the algorithms for the application of medical diagnosis. Results display that the proposed hybrid training algorithm, significantly outperforms the traditional training algorithm, by enhancing the accuracies of the ANNs with an increase of 20% in the average testing accuracy and 0.7% increase in the best testing accuracy.