Search

Training generative models like Generative Adversarial Network (GAN) is challenging for noisy data. A novel curriculum learning algorithm pertaining to clustering is proposed to address this issue in this paper. The curriculum construction is based on the centrality of underlying clusters in data points. The data points of high centrality takes priority of being fed into generative models during training. To make our algorithm scalable to large-scale data, the active set is devised, in the sense that every round of training proceeds only on an active subset containing a small fraction of already trained data and the incremental data of lower centrality. Moreover, the geometric analysis is presented to interpret the necessity of cluster curriculum for generative models. The experiments on cat and human-face data validate that our algorithm is able to learn the optimal generative models (e.g. ProGAN) with respect to specified quality metrics for noisy data. An interesting finding is that the optimal cluster curriculum is closely related to the critical point of the geometric percolation process formulated in the paper.

Nowadays China are experiencing the progress of urbanization (urban population explosion, the increase of school-age children) which makes it necessary for the urgent reform in education facilities locate in downtown. This paper discusses the optimized design of planning and building reconstruction of primary and secondary school in downtown by analyzing the school status. We must inherit the history of old schoolyard should be inherited to keep harmonious landscape as well as apply the modern technology to complete the function and create sufficient space in order to inject new vitality into the traditional schoolyard. The feasibility and effect of optimal design ideas together with design practice are also analyzed in this paper.