Mountains aren't just beautiful: these locales also tend to host some of the richest diversity of species on the planet. We've known this for a long time--ever since Alexander von Humboldt, the Prussian geographer and naturalist, first climbed up the Andes in the 18th century. But nobody has really figured out why. One popular hypothesis goes like this: the reason why mountains have so many different species is that, as mountains are uplifted by colliding tectonic plates, the process creates more environments, and therefore more opportunities for new species to adapt to them. However, this hypothesis never had any explicit quantitative testing until now, according to a recent study published in the Proceedings of the National Academy of Sciences.
The need for diversification of recommendation lists manifests in a number of recommender systems use cases. However, an increase in diversity may undermine the utility of the recommendations, as relevant items in the list may be replaced by more diverse ones. In this work we propose a novel method for maximizing the utility of the recommended items subject to the diversity of user's tastes, and show that an optimal solution to this problem can be found greedily. We evaluate the proposed method in two online user studies as well as in an offline analysis incorporating a number of evaluation metrics. The results of evaluations show the superiority of our method over a number of baselines.
Ensemble methods, which train multiple learners for a task, are among the state-of-the-art learning approaches. The diversity of the component learners has been recognized as a key to a good ensemble, and existing ensemble methods try different ways to encourage diversity, mostly by heuristics. In this paper, we propose the diversity regularized machine (DRM) in a mathematical programming framework, which efficiently generates an ensemble of diverse support vector machines (SVMs). Theoretical analysis discloses that the diversity constraint used in DRM can lead to an effective reduction on its hypothesis space complexity, implying that the diversity control in ensemble methods indeed plays a role of regularization as in popular statistical learning approaches. Experiments show that DRM can significantly improve generalization ability and is superior to some state-of-the-art SVM ensemble methods.
Debate is open as to whether social media communities resemble real-life communities, and to what extent. We contribute to this discussion by testing whether established sociological theories of real-life networks hold in Twitter. In particular, for 228,359 Twitter profiles, we compute network metrics (e.g., reciprocity, structural holes, simmelian ties) that the sociological literature has found to be related to parts of one's social world (i.e., to topics, geography and emotions), and test whether these real-life associations still hold in Twitter. We find that, much like individuals in real-life communities, social brokers (those who span structural holes) are opinion leaders who tweet about diverse topics, have geographically wide networks, and express not only positive but also negative emotions. Furthermore, Twitter users who express positive (negative) emotions cluster together, to the extent of having a correlation coefficient between one's emotions and those of friends as high as 0.45. Understanding Twitter's social dynamics does not only have theoretical implications for studies of social networks but also has practical implications, including the design of self-reflecting user interfaces that make people aware of their emotions, spam detection tools, and effective marketing campaigns.
Standard practice in training neural networks involves initializing the weights in an independent fashion. The results of recent work suggest that feature "diversity" at initialization plays an important role in training the network. However, other initialization schemes with reduced feature diversity have also been shown to be viable. In this work, we conduct a series of experiments aimed at elucidating the importance of feature diversity at initialization. We show that a complete lack of diversity is harmful to training, but its effects can be counteracted by a relatively small addition of noise - even the noise in standard non-deterministic GPU computations is sufficient. Furthermore, we construct a deep convolutional network with identical features at initialization and almost all of the weights initialized at 0 that can be trained to reach accuracy matching its standard-initialized counterpart.