No abstract uploaded!

We discuss quantum graphs consisting of a compact part and semi-infinite leads. Such a system may have embedded eigenvalues if some edge lengths in the compact part are rationally related. If such a relation is perturbed, these eigenvalues may turn into resonances; we analyze this effect both generally and in simple examples.

The longstanding open problem of approximating all singular vertex couplings in a quantum graph is solved. We present a construction in which the edges are decoupled; an each pair of their endpoints is joined by an edge carrying a potential and a vector potential coupled to the loose edges by a coupling. It is shown that if the lengths of the connecting edges shrink to zero and the potentials are properly scaled, the limit can yield any prescribed singular vertex coupling, and moreover, that such an approximation converges in the norm-resolvent sense.

We discuss the discrete spectrum of N particles in a curved planar waveguide. If they are neutral fermions, the maximum number of particles that the waveguide can bind is given by a one-particle BirmanSchwinger bound in combination with the Pauli principle. On the other hand, if they are charged, e.g., electrons in a bent quantum wire, the Coulomb repulsion plays a crucial role. We prove a sufficient condition under which the discrete spectrum of such a system is empty.

Remarkable achievements have been attained by deep neural networks in various applications. However, the increasing depth and width of such models also lead to explosive growth in both storage and computation, which has restricted the deployment of deep neural networks on resource-limited edge devices. To address this problem, we propose the so-called SCAN framework for networks training and inference, which is orthogonal and complementary to existing acceleration and compression methods. The proposed SCAN firstly divides neural networks into multiple sections according to their depth and constructs shallow classifiers upon the intermediate features of different sections. Moreover, attention modules and knowledge distillation are utilized to enhance the accuracy of shallow classifiers. Based on this architecture, we further propose a threshold controlled scalable inference mechanism to approach human-like sample-specific inference. Experimental results show that SCAN can be easily equipped on various neural networks without any adjustment on hyper-parameters or neural networks architectures, yielding significant performance gain on CIFAR100 and ImageNet. Codes are be released on https://github.com/ArchipLab-LinfengZhang/pytorch-scalable-neural-networks.