Wenxu Wang

Andrey Ignatov, Radu Timofte, Cheng-Ming Chiang et al.

Video super-resolution is one of the most popular tasks on mobile devices, being widely used for an automatic improvement of low-bitrate and low-resolution video streams. While numerous solutions have been proposed for this problem, they are usually quite computationally demanding, demonstrating low FPS rates and power efficiency on mobile devices. In this Mobile AI challenge, we address this problem and propose the participants to design an end-to-end real-time video super-resolution solution for mobile NPUs optimized for low energy consumption. The participants were provided with the REDS training dataset containing video sequences for a 4X video upscaling task. The runtime and power efficiency of all models was evaluated on the powerful MediaTek Dimensity 9000 platform with a dedicated AI processing unit capable of accelerating floating-point and quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 500 FPS rate and 0.2 [Watt / 30 FPS] power consumption. A detailed description of all models developed in the challenge is provided in this paper.

Jing Wang, Wonho Bae, Jiahong Chen et al.

Recent work on latent diffusion models (LDMs) has focused almost exclusively on generative tasks, leaving their potential for discriminative transfer largely unexplored. We introduce Discriminative Vicinity Diffusion (DVD), a novel LDM-based framework for a more practical variant of source-free domain adaptation (SFDA): the source provider may share not only a pre-trained classifier but also an auxiliary latent diffusion module, trained once on the source data and never exposing raw source samples. DVD encodes each source feature's label information into its latent vicinity by fitting a Gaussian prior over its k-nearest neighbors and training the diffusion network to drift noisy samples back to label-consistent representations. During adaptation, we sample from each target feature's latent vicinity, apply the frozen diffusion module to generate source-like cues, and use a simple InfoNCE loss to align the target encoder to these cues, explicitly transferring decision boundaries without source access. Across standard SFDA benchmarks, DVD outperforms state-of-the-art methods. We further show that the same latent diffusion module enhances the source classifier's accuracy on in-domain data and boosts performance in supervised classification and domain generalization experiments. DVD thus reinterprets LDMs as practical, privacy-preserving bridges for explicit knowledge transfer, addressing a core challenge in source-free domain adaptation that prior methods have yet to solve.

Le-Zhi Wang, Ri-Qi Su, Zi-Gang Huang et al.

In spite of the recent interest and advances in linear controllability of complex networks, controlling nonlinear network dynamics remains to be an outstanding problem. We develop an experimentally feasible control framework for nonlinear dynamical networks that exhibit multistability (multiple coexisting final states or attractors), which are representative of, e.g., gene regulatory networks (GRNs). The control objective is to apply parameter perturbation to drive the system from one attractor to another, assuming that the former is undesired and the latter is desired. To make our framework practically useful, we consider RESTRICTED parameter perturbation by imposing the following two constraints: (a) it must be experimentally realizable and (b) it is applied only temporarily. We introduce the concept of ATTRACTOR NETWORK, in which the nodes are the distinct attractors of the system, and there is a directional link from one attractor to another if the system can be driven from the former to the latter using restricted control perturbation. Introduction of the attractor network allows us to formulate a controllability framework for nonlinear dynamical networks: a network is more controllable if the underlying attractor network is more strongly connected, which can be quantified. We demonstrate our control framework using examples from various models of experimental GRNs. A finding is that, due to nonlinearity, noise can counter-intuitively facilitate control of the network dynamics.

Yu-Zhong Chen, Lezhi Wang, Wenxu Wang et al.

In this paper, we investigate the linear controllability framework for complex networks from a physical point of view. There are three main results. (1) If one applies control signals as determined from the structural controllability theory, there is a high probability that the control energy will diverge. Especially, if a network is deemed controllable using a single driving signal, then most likely the energy will diverge. (2) The energy required for control exhibits a power-law scaling behavior. (3) Applying additional control signals at proper nodes in the network can reduce and optimize the energy cost. We identify the fundamental structures embedded in the network, the longest control chains, which determine the control energy and give rise to the power-scaling behavior. (To our knowledge, this was not reported in any previous work on control of complex networks.) In addition, the issue of control precision is addressed. These results are supported by extensive simulations from model and real networks, physical reasoning, and mathematical analyses. Notes on the submission history of this work: This work started in late 2012. The phenomena of power-law energy scaling and energy divergence with a single controller were discovered in 2013. Strategies to reduce and optimize control energy was articulated and tested in 2013. The senior co-author (YCL) gave talks about these results at several conferences, including the NETSCI 2014 Satellite entitled "Controlling Complex Networks" on June 2, 2014. The paper was submitted to PNAS in September 2014 and was turned down. It was revised and submitted to PRX in early 2015 and was rejected. After that it was revised and submitted to Nature Communications in May 2015 and again was turned down.