Beihang University

About: Beihang University is a education organization based out in Beijing, China. It is known for research contribution in the topics: Control theory & Microstructure. The organization has 67002 authors who have published 73507 publications receiving 975691 citations. The organization is also known as: Beijing University of Aeronautics and Astronautics.

Accelerating convolutional networks via global & dynamic filter pruning

Abstract: Accelerating convolutional neural networks has recently received ever-increasing research focus. Among various approaches proposed in the literature, filter pruning has been regarded as a promising solution, which is due to its advantage in significant speedup and memory reduction of both network model and intermediate feature maps. To this end, most approaches tend to prune filters in a layerwise fixed manner, which is incapable to dynamically recover the previously removed filter, as well as jointly optimize the pruned network across layers. In this paper, we propose a novel global & dynamic pruning (GDP) scheme to prune redundant filters for CNN acceleration. In particular, GDP first globally prunes the unsalient filters across all layers by proposing a global discriminative function based on prior knowledge of each filter. Second, it dynamically updates the filter saliency all over the pruned sparse network, and then recovers the mistakenly pruned filter, followed by a retraining phase to improve the model accuracy. Specially, we effectively solve the corresponding nonconvex optimization problem of the proposed GDP via stochastic gradient descent with greedy alternative updating. Extensive experiments show that the proposed approach achieves superior performance to accelerate several cutting-edge CNNs on the ILSVRC 2012 benchmark, comparing to the state-of-the-art filter pruning methods.

Exact phase transitions in random constraint satisfaction problems

Abstract: In this paper we propose a new type of random CSP model, called Model RB, which is a revision to the standard Model B. It is proved that phase transitions from a region where almost all problems are satisfiable to a region where almost all problems are unsatisfiable do exist for Model RB as the number of variables approaches infinity. Moreover, the critical values at which the phase transitions occur are also known exactly. By relating the hardness of Model RB to Model B, it is shown that there exist a lot of hard instances in Model RB.

Generating Defect-Rich Bismuth for Enhancing the Rate of Nitrogen Electroreduction to Ammonia

Abstract: The electrochemical N2 fixation, which is far from practical application in aqueous solution under ambient conditions, is extremely challenging and requires a rational design of electrocatalytic centers. We observed that bismuth (Bi) might be a promising candidate for this task because of its weak binding with H adatoms, which increases the selectivity and production rate. Furthermore, we successfully synthesized defect-rich Bi nanoplates as an efficient noble-metal-free N2 reduction electrocatalyst via a low-temperature plasma bombardment approach. When exclusively using 1 H NMR measurements with N2 gas as a quantitative testing method, the defect-rich Bi(110) nanoplates achieved a 15 NH3 production rate of 5.453 μg mgBi -1 h-1 and a Faradaic efficiency of 11.68 % at -0.6 V vs. RHE in aqueous solution at ambient conditions.