Search

In this paper, we focus on addressing the open-set face identification problem on a few-shot gallery by fine-tuning. The problem assumes a realistic scenario for face identification, where only a small number of face images is given for enrollment and any unknown identity must be rejected during identification. We observe that face recognition models pretrained on a large dataset and naively fine-tuned models perform poorly for this task. Motivated by this issue, we propose an effective fine-tuning scheme with classifier weight imprinting and exclusive BatchNorm layer tuning. For further improvement of rejection accuracy on unknown identities, we propose a novel matcher called Neighborhood Aware Cosine (NAC) that computes similarity based on neighborhood information. We validate the effectiveness of the proposed schemes thoroughly on large-scale face benchmarks across different convolutional neural network architectures. The source code for this project is available at: https://github.com/1ho0jin1/OSFI-by-FineTuning

Compartment models are widely used in climate science, epidemiology, and physics, among other disciplines. An important example of a compartment model is susceptible-infected-recovered (SIR) model, which can describe disease dynamics. Bayesian inference for SIR models is challenging because likelihood evaluation requires solving expensive ordinary differential equations. Although variational inference (VI) can be a fast alternative to the traditional Bayes approaches, VI has limited applicability due to boundary issues and local optima problems. To address these challenges, we propose flexible VI methods based on deep generative models that do not require parametric assumptions on the variational distribution. We embed a surjective transformation in our framework to avoid posterior truncation at the boundary. We provide theoretical conditions that guarantee the success of the algorithm. Furthermore, our temperature annealing scheme can avoid being trapped in local optima through a series of intermediate posteriors. We apply our method to variants of SIR models, illustrating that the proposed method can provide fast and accurate inference compared to its competitors.

Given the huge volume of cross-border flows, effective and efficient control of trade becomes more crucial in protecting people and society from illicit trade. However, limited accessibility of the transaction-level trade datasets hinders the progress of open research, and lots of customs administrations have not benefited from the recent progress in data-based risk management. In this paper, we introduce an import declaration dataset to facilitate the collaboration between domain experts in customs administrations and researchers from diverse domains, such as data science and machine learning. The dataset contains 54,000 artificially generated trades with 22 key attributes, and it is synthesized with conditional tabular GAN while maintaining correlated features. Synthetic data has several advantages. First, releasing the dataset is free from restrictions that do not allow disclosing the original import data. The fabrication step minimizes the possible identity risk which may exist in trade statistics. Second, the published data follow a similar distribution to the source data so that it can be used in various downstream tasks. Hence, our dataset can be used as a benchmark for testing the performance of any classification algorithm. With the provision of data and its generation process, we open baseline codes for fraud detection tasks, as we empirically show that more advanced algorithms can better detect fraud., Comment: Datasets: https://github.com/Seondong/Customs-Declaration-Datasets

Convolutional neural networks (CNNs) provide flexible function approximations for a wide variety of applications when the input variables are in the form of images or spatial data. Although CNNs often outperform traditional statistical models in prediction accuracy, statistical inference, such as estimating the effects of covariates and quantifying the prediction uncertainty, is not trivial due to the highly complicated model structure and overparameterization. To address this challenge, we propose a new Bayes approach by embedding CNNs within the generalized linear model (GLM) framework. We use extracted nodes from the last hidden layer of CNN with Monte Carlo dropout as informative covariates in GLM. This improves accuracy in prediction and regression coefficient inference, allowing for the interpretation of coefficient and uncertainty quantification. By fitting ensemble GLMs across multiple realizations from Monte Carlo dropout, we can fully account for uncertainties in model estimation. We apply our methods to simulated and real data examples, including non-Gaussian spatial data, brain tumor image data, and fMRI data. The algorithm can be broadly applicable to image regressions or correlated data analysis by enabling accurate Bayesian inference quickly., Comment: 25 pages, 7 figures

Periocular biometric, the peripheral area of the ocular, is a collaborative alternative to the face, especially when the face is occluded or masked. However, in practice, sole periocular biometric capture the least salient facial features, thereby lacking discriminative information, particularly in wild environments. To address these problems, we transfer discriminatory information from the face to support the training of a periocular network by using knowledge distillation. Specifically, we leverage face images for periocular embedding learning, but periocular alone is utilized for identity identification or verification. To enhance periocular embeddings by face effectively, we proposeConsistent Knowledge Distillation (CKD) that imposes consistency between face and periocular networks across prediction and feature layers. We find that imposing consistency at the prediction layer enables (1) extraction of global discriminative relationship information from face images and (2) effective transfer of the information from the face network to the periocular network. Particularly, consistency regularizes the prediction units to extract and store profound inter-class relationship information of face images. (3) The feature layer consistency, on the other hand, makes the periocular features robust against identity-irrelevant attributes. Overall, CKD empowers the sole periocular network to produce robust discriminative embeddings for periocular recognition in the wild. We theoretically and empirically validate the core principles of the distillation mechanism in CKD, discovering that CKD is equivalent to label smoothing with a novel sparsity-oriented regularizer that helps the network prediction to capture the global discriminative relationship. Extensive experiments reveal that CKD achieves state-of-the-art results on standard periocular recognition benchmark datasets., Comment: Submitted to IEEE TIP. The first two authors have contributed equally

Nonstationary non-Gaussian spatial data are common in many disciplines, including climate science, ecology, epidemiology, and social sciences. Examples include count data on disease incidence and binary satellite data on cloud mask (cloud/no-cloud). Modeling such datasets as stationary spatial processes can be unrealistic since they are collected over large heterogeneous domains (i.e., spatial behavior differs across subregions). Although several approaches have been developed for nonstationary spatial models, these have focused primarily on Gaussian responses. In addition, fitting nonstationary models for large non-Gaussian datasets is computationally prohibitive. To address these challenges, we propose a scalable algorithm for modeling such data by leveraging parallel computing in modern high-performance computing systems. We partition the spatial domain into disjoint subregions and fit locally nonstationary models using a carefully curated set of spatial basis functions. Then, we combine the local processes using a novel neighbor-based weighting scheme. Our approach scales well to massive datasets (e.g., 1 million samples) and can be implemented in nimble, a popular software environment for Bayesian hierarchical modeling. We demonstrate our method to simulated examples and two large real-world datasets pertaining to infectious diseases and remote sensing.