Particle Competition and Cooperation in Networks for Semi-Supervised Learning

Semi-supervised learning is one of the important topics in machine learning, concerning with pattern classification where only a small subset of data is labeled. In this paper, a new network-based (or graph-based) semi-supervised classification model is proposed. It employs a combined random-greedy walk of particles, with competition and cooperation mechanisms, to propagate class labels to the whole network. Due to the competition mechanism, the proposed model has a local label spreading fashion, i.e., each particle only visits a portion of nodes potentially belonging to it, while it is not allowed to visit those nodes definitely occupied by particles of other classes. In this way, a "divide-and-conquer" effect is naturally embedded in the model. As a result, the proposed model can achieve a good classification rate while exhibiting low computational complexity order in comparison to other network-based semi-supervised algorithms. Computer simulations carried out for synthetic and real-world data sets provide a numeric quantification of the performance of the method.

Selectivity and Mechanisms Driven by Reaction Dynamics: The Case of the Gas-Phase OH-+CH3ONO2 Reaction

Well-established statistical approaches such as transition-state theory based on high-level calculated potential energy profiles are unable to account for the selectivity observed in the gas-phase OH- + CH3ONO2 reaction. This reaction can undergo bimolecular nucleophilic displacement at either the carbon center (S(N)2@C) or the nitrogen center (S(N)2@N) as well as a proton abstraction followed by dissociation (E(CO)2) pathway. Direct dynamics simulations yield an S(N)2:E(CO)2 product ratio in close agreement with experiment and show that the lack of reactivity at the nitrogen atom is due to the highly negative electrostatic potential generated by the oxygen atoms in the ONO2 group that scatters the incoming OH-. In addition to these dynamical effects, the nonstatistical behavior of these reactions is attributed to the absence of equilibrated reactant complexes and to the large number of recrossings, which might be present in several ion-molecule gas-phase reactions.