We propose to detect Deepfake generated by face manipulation based on one of their fundamental features: images are blended by patches from multiple sources, carrying distinct and persistent source features. In particular, we propose a novel representation learning approach for this task, called patch-wise consistency learning (PCL). It learns by measuring the consistency of image source features, resulting to representation with good interpretability and robustness to multiple forgery methods. We develop an inconsistency image generator (I2G) to generate training data for PCL and boost its robustness. We evaluate our approach on seven popular Deepfake detection datasets. Our model achieves superior detection accuracy and generalizes well to unseen generation methods. On average, our model outperforms the state-of-the-art in terms of AUC by 2% and 8% in the in-and cross-dataset evaluation, respectively.

We propose a simple yet highly effective method that addresses the mode-collapse problem in the Conditional Generative Adversarial Network (cGAN). Although conditional distributions are multi-modal (i.e., having many modes) in practice, most cGAN approaches tend to learn an overly simplified distribution where an input is always mapped to a single output regardless of variations in latent code. To address such issue, we propose to explicitly regularize the generator to produce diverse outputs depending on latent codes. The proposed regularization is simple, general, and can be easily integrated into most conditional GAN objectives. Additionally, explicit regularization on generator allows our method to control a balance between visual quality and diversity. We demonstrate the effectiveness of our method on three conditional generation tasks: image-to-image translation, image inpainting, and future video prediction. We show that simple addition of our regularization to existing models leads to surprisingly diverse generations, substantially outperforming the previous approaches for multi-modal conditional generation specifically designed in each individual task.

No abstract

No abstract

Metallic glasses (MGs), with high density of low coordination sites and high Gibbs free energy state, are novel promising and competitive candidates in the family of electrochemical catalysts. However, it remains a grand challenge to modify the properties of MGs by control of the disordered atomic structure. Recently, nanostructured metallic glasses (NGs), consisting of amorphous nanometer‐sized grains connected by amorphous interfaces, have been reported to exhibit tunable properties compared to the MGs with identical chemical composition. Here, it is demonstrated that electrodeposited Ni–P NG is characterized by an extremely high energy state due to its heterogeneous structure, which significantly promotes the catalytic performance. Moreover, the Ni–P NG with a heterogeneous structure is a perfect precursor for the fabrication of unique honey‐like nanoporous structure, which displays superior catalytic performance in the urea oxidation reaction (UOR). Specifically, modified Ni–P NG requires a potential of mere 1.36 V at 10 mA cm–2, with a Tafel slope of 13 mV dec–1, which is the best UOR performance in Ni‐based alloys. The present work demonstrates that the nanostructurization of MGs provides a universal and effective pathway to upgrade the energy state of MGs for the design of high‐performance catalysts in energy conversion.

This work proposes a novel Graph-based neural ArchiTecture Encoding Scheme, a.k.a. GATES, to improve the predictor-based neural architecture search. Specifically, different from existing graphbased schemes, GATES models the operations as the transformation of the propagating information, which mimics the actual data processing of neural architecture. GATES is a more reasonable modeling of the neural architectures, and can encode architectures from both the "operation on node" and "operation on edge" cell search spaces consistently. Experimental results on various search spaces confirm GATES's effectiveness in improving the performance predictor. Furthermore, equipped with the improved performance predictor, the sample efficiency of the predictorbased neural architecture search (NAS) flow is boosted.

A notion of quantum natural evolution strategies is introduced, which provides a geometric synthesis of a number of known quantum/classical algorithms for performing classical black-box optimization. The recent work of Gomes et al (2019 arXiv:1910.10675) on heuristic combinatorial optimization using neural quantum states is pedagogically reviewed in this context, emphasizing the connection with natural evolution strategies (NES). The algorithmic framework is illustrated for approximate combinatorial optimization problems, and a systematic strategy is found for improving the approximation ratios. In particular, it is found that NES can achieve approximation ratios competitive with widely used heuristic algorithms for Max-Cut, at the expense of increased computation time.

No abstract