Feature representation of data is a key issue for recognition related tasks. Inspired by the creative ability of human beings, in this paper we propose a novel feature learning framework named within-class reconstruction (WCR). In WCR, the feature representation of the input sample are used to reconstruct all the samples within the same class. We minimize the mean squared error (MSE) cost function to update feature extracting functions. Furthermore, most unsupervised learning methods such as auto-encoders could embed in the proposed framework. To evaluate the effectiveness of the proposed framework, CNN is used to extract the feature representations and reconstruct the within-class samples. The experimental results demonstrate that the representations learned by the proposed WCR achieve better performance than that of auto-encoders. All the codes have been made publicly available at https://github.com/step123456789/wcr.

As the development of collaborative applications, the parallel interactions among processes are more and more complicated and frequent. However, modeling of the interactions among many collaborative processes is a complicated and error-prone procedure. In this paper, firstly, a novel model based on Petri net, called PIPN, was proposed. PIPN is suitable to define and analyze the parallel interactions among collaborative processes. Secondly, seven parallel interactive modes were summarized according to three views of parallel interactions, which are unidirectional or bidirectional, single-point or multi-point, and synchronous or asynchronous. Then the formal definitions and control flow graphs of these modes were given. Finally, an example, called micro blog, was modeled to verify the reasonableness and feasibility of the work in this paper.