Federated learning (FL) is a promising approach for enhancing data privacy preservation, particularly for authentication systems. However, limited round communications, scarce representation, and scalability pose significant challenges to its deployment, hindering its full potential. In this paper, we propose ‘ProtoFL’, Prototypical Representation Distillation based unsupervised Federated Learning to enhance the representation power of a global model and reduce round communication costs. Additionally, we introduce a local one-class classifier based on normalizing flows to improve performance with limited data. Our study represents the first investigation of using FL to improve one-class classification performance. We conduct extensive experiments on five widely used benchmarks, namely MNIST, CIFAR-10, CIFAR-100, ImageNet-30, and Keystroke-Dynamics, to demonstrate the superior performance of our proposed framework over previous methods in the literature.

This paper considers classification problems with hierarchically organized classes. We force the classifier (hyperplane) of each class to belong to a sphere manifold, whose center is the classifier of its super-class. Then, individual sphere manifolds are connected based on their hierarchical relations. Our technique replaces the last layer of a neural network by combining a spherical fully-connected layer with a hierarchical layer. This regularization is shown to improve the performance of widely used deep neural network architectures (ResNet and DenseNet) on publicly available datasets (CIFAR100, CUB200, Stanford dogs, Stanford cars, and Tiny-ImageNet).

While humans can solve a visual puzzle that requires logical reasoning by observing only few samples, it would require training over a large number of samples for state-of-the-art deep reasoning models to obtain similar performance on the same task. In this work, we propose to solve such a few-shot (or low-shot) abstract visual reasoning problem by resorting to \emph{analogical reasoning}, which is a unique human ability to identify structural or relational similarity between two sets. Specifically, we construct analogical and non-analogical training pairs of two different problem instances, e.g., the latter is created by perturbing or shuffling the original (former) problem. Then, we extract the structural relations among elements in both domains in a pair by enforcing analogical ones to be as similar as possible, while minimizing similarities between non-analogical ones. This analogical contrastive learning allows to effectively learn the relational representations of given abstract reasoning tasks. We validate our method on RAVEN dataset, on which it outperforms state-of-the-art method, with larger gains when the training data is scarce. We further meta-learn our analogical contrastive learning model over the same tasks with diverse attributes, and show that it generalizes to the same visual reasoning problem with unseen attributes.

Large-scale simulation codes that model complicated science and engineering applications typically have huge and complex code bases. For such simulation codes, where bit-for-bit comparisons are too restrictive, finding the source of statistically significant discrepancies (e.g., from a previous version, alternative hardware or supporting software stack) in output is non-trivial at best. Although there are many tools for program comprehension through debugging or slicing, few (if any) scale to a model as large as the Community Earth System Model (CESM#8482;), which consists of more than 1.5 million lines of Fortran code. Currently for the CESM, we can easily determine whether a discrepancy exists in the output using a by now well-established statistical consistency testing tool. However, this tool provides no information as to the possible cause of the detected discrepancy, leaving developers in a seemingly impossible (and frustrating) situation. Therefore, our aim in this work is to provide the tools to enable developers to trace a problem detected through the CESM output to its source. To this end, our strategy is to reduce the search space for the root cause(s) to a tractable size via a series of techniques that include creating a directed graph of internal CESM variables, extracting a subgraph (using a form of hybrid program slicing), partitioning into communities, and ranking nodes by centrality. Runtime variable sampling then becomes feasible in this reduced search space. We demonstrate the utility of this process on multiple examples of CESM simulation output by illustrating how sampling can be performed as part of an efficient parallel iterative refinement procedure to locate error sources, including sensitivity to CPU instructions. By providing CESM developers with tools to identify and understand the reason for statistically distinct output, we have positively impacted the CESM software development cycle and, in particular, its focus on quality assurance.

Face Parsing assigns every pixel in a facial image with a semantic label, which could be applied in various applications including face recognition, facial beautification, affective computing and animation. While lots of progress have been made in this field, current state-of-the-art methods still fail to extract real effective feature and restore accurate score map, especially for those facial parts which have large variations of deformation and fairly similar appearance, e.g. mouth, eyes and thin eyebrows. In this paper, we propose a novel pixel-wise face parsing method called Residual Encoder Decoder Network (RED-Net), which combines a feature-rich encoder-decoder framework with adaptive prior mechanism. Our encoder-decoder framework extracts feature with ResNet and decodes the feature by elaborately fusing the residual architectures in to deconvolution. This framework learns more effective feature comparing to that learnt by decoding with interpolation or classic deconvolution operations. To overcome the appearance ambiguity between facial parts, an adaptive prior mechanism is proposed in term of the decoder prediction confidence, allowing refining the final result. The experimental results on two public datasets demonstrate that our method outperforms the state-of-the-arts significantly, achieving improvements of F-measure from 0.854 to 0.905 on Helen dataset, and pixel accuracy from 95.12% to 97.59% on the LFW dataset. In particular, convincing qualitative examples show that our method parses eye, eyebrow, and lip regins more accurately.

Accurate age estimation from a facial image is quite challenging, since physical age and apparent age can be quite different, and this difference is dependent on gender, ethnicity, and many other factors. Multitask deep learning is one of the approach to improve age estimation by employing auxiliary tasks, such as gender recognition, that are related to the primary task. However, in traditional multitask learning for age estimation, the relationship between the primary and auxiliary tasks is difficult to describe; how the auxiliary tasks enhance the model for the primary objective is ambiguous. In this letter, we propose a conditional multitask learning method that architecturally factorizes an age variable into gender-conditioned age probabilities in a deep neural network. The lack of accurate training labels with discrete age values is another critical limitation to training age estimation models. Therefore, we propose a label expansion method that increases the number of accurate labels from weakly supervised categorical labels. To verify the generality of the proposed method, we perform intensive experiments on the publicly available MORPH-II and FG-NET datasets. The proposed methods outperform state-of-the art methods in both age estimation and gender recognition accuracy. These performance gains are verified on well-known deep network architectures-VGG-16, CASIA-WebFace, and Alexnet-to confirm the proposed methods generality.

A kernel or mini-app is a self-contained small application that retains certain characteristics of the original application [7]. Working on a kernel or mini-app in the place of the original application can dramatically reduce the resources and effort required for performing software tasks such as performance optimization and porting to new platforms. However, using kernel as a proxy is based on the assumption that it represents the original application in the context of how it is being used. In this paper, we introduce an extension to the Fortran Kernel Generator (KGen) which is an automated kernel extraction tool [1]. The extension allows comparison of the execution characteristics between the original application and the generated kernel using descriptive statistics. From the comparison, the user is provided with statistics that provide information on the degree and context of representativeness of the kernel. KGen also utilizes the information generated to help it to automatically improve representativeness of the kernels whilst reducing the size of the workload generated. We applied this extension to three kernels. One is generated from a Fortran scientific library and the remaining two are generated from an earth system model. We have demonstrated that the descriptive statistics provided in the enhancement provide not only quantitative metrics and context of representativeness but also a way to improve the quality of representativeness of the kernels generated.

In this letter, we propose a novel method for background subtraction in outdoor scenes. Inspired by the observation that the orthogonal decomposition onto a set of pixel intensities efficiently reveals illumination effects, we exploit a simple, yet powerful feature for describing the underlying structure of the local region in a given video, the so-called illumination-invariant structural complexity (IISC). In contrast to previous approaches still suffering from high-level false positives driven by varying illuminations in outdoor environments, our IISC feature has an ability to greatly discriminate structural changes by moving objects from those by illumination effects. We also provide the theoretical analysis to confirm that the proposed IISC feature is useful for modeling the background under diverse lighting conditions. Moreover, our framework does not require any preprocessing task. Experimental results on various datasets demonstrate that the proposed method is effective for video surveillance in a wide range of outdoor environments.

In transport modeling and prediction, trip purposes play an important role. The most particular case is activity-based modeling, whereby mobility choices (e.g. mode, path, and departure time) are made in order to carry out specific activities. A current challenge, however, lies on getting appropriate data that relates observed trips with their purpose. Recently, a Smartphone-based travel survey (the Future Mobility Survey, FMS) was conducted in Singapore that collected location data from 793 participants. Each FMS user was required to collect data for at least 14 days and validate at least 5 of them. This dataset presents diverse opportunities in terms of developing machine learning models for the future versions of FMS, where the validation process is intelligent and easy to use (e.g. having pre-filled activities associated to the user traces). This paper proposes a learning model that, given a stop location, identifies the most likely activity associated to it. Our data often contains errors or noise due to limited functionality of physical sensors in a dense area, and human mistakes in the validation process. To alleviate this effect, we generate heterogeneous features by different spatial quantization techniques and apply ensemble learning for a good generalization performance.

Abstract This paper presents a two‐stage procedure to combine multiple face traits for identity authentication. At the first stage, a high dimensional random projection is applied to the raw visual and infrared face images to extract useful information relevant to each identity. This is followed by a dimension reduction using eigenfeature regularization and extraction (ERE). At the second stage, the scores from two verification systems based on each face modality are fused by an error minimization algorithm. This error minimization algorithm directly optimizes the verification accuracy by adjusting the parameters of a polynomial classifier. Two data sets consisting of visual and infrared face images have been used for experimentation. Our empirical observation shows encouraging results regarding the effectiveness of the proposed method. Copyright © 2011 John Wiley & Sons, Ltd.

This paper presents a score level fusion of visual and infrared face image verification systems. A high dimensional random projection is first applied to the raw visual and infrared face images to extract useful information relevant to each identity. This is followed by a dimension reduction using eigenfeature regularization and extraction. The resultant templates are then compared for decision scores generation. Finally the scores from the visual and infrared face image verification systems are fused by an error rate minimization formulation. Our empirical observation shows encouraging results regarding the effectiveness of the fusion.