arXiv subjects

Chong Zhang

Publications and source records attributed to Chong Zhang.

At least 163 records · Page 9Linked to original sources

arXiv2020

Domain Adaptation Gaze Estimation by Embedding with Prediction Consistency

Gaze is the essential manifestation of human attention. In recent years, a series of work has achieved high accuracy in gaze estimation. However, the inter-personal difference limits the reduction of the subject-independent gaze estimation error. This paper proposes an unsupervised method for domain adaptation gaze estimation to eliminate the impact of inter-personal diversity. In domain adaption, we design an embedding representation with prediction consistency to ensure that the linear relationship between gaze directions in different domains remains consistent on gaze space and embedding space. Specifically, we employ source gaze to form a locally linear representation in the gaze space for each target domain prediction. Then the same linear combinations are applied in the embedding space to generate hypothesis embedding for the target domain sample, remaining prediction consistency. The deviation between the target and source domain is reduced by approximating the predicted and hypothesis embedding for the target domain sample. Guided by the proposed strategy, we design Domain Adaptation Gaze Estimation Network(DAGEN), which learns embedding with prediction consistency and achieves state-of-the-art results on both the MPIIGaze and the EYEDIAP datasets.

cs.CV↗

arXiv2020

An Efficient Adversarial Attack for Tree Ensembles

We study the problem of efficient adversarial attacks on tree based ensembles such as gradient boosting decision trees (GBDTs) and random forests (RFs). Since these models are non-continuous step functions and gradient does not exist, most existing efficient adversarial attacks are not applicable. Although decision-based black-box attacks can be applied, they cannot utilize the special structure of trees. In our work, we transform the attack problem into a discrete search problem specially designed for tree ensembles, where the goal is to find a valid "leaf tuple" that leads to mis-classification while having the shortest distance to the original input. With this formulation, we show that a simple yet effective greedy algorithm can be applied to iteratively optimize the adversarial example by moving the leaf tuple to its neighborhood within hamming distance 1. Experimental results on several large GBDT and RF models with up to hundreds of trees demonstrate that our method can be thousands of times faster than the previous mixed-integer linear programming (MILP) based approach, while also providing smaller (better) adversarial examples than decision-based black-box attacks on general $\ell_p$ ($p=1, 2, \infty$) norm perturbations. Our code is available at https://github.com/chong-z/tree-ensemble-attack.

cs.LG↗

arXiv2020

Parameter estimation in FACS-seq enables high-throughput characterization of phenotypic heterogeneity

Phenotypic heterogeneity is a most fascinating property of a population of cells, which shows the differences among individuals even with the same genetic background and extracellular environmental conditions. However, the lack of high-throughput analysis of phenotypic diversity has limited our research progress. To deal with it, we constructed a novel parameter estimation method in FACS-seq, a commonly used experimental framework, to achieve simultaneous characterization of thousands of variants in a library. We further demonstrated the model's ability in estimating the expression properties of each variant, which we believe can help to decipher the mechanisms of phenotypic heterogeneity.

q-bio.QM↗

arXiv2020

Learning End-to-End Action Interaction by Paired-Embedding Data Augmentation

In recognition-based action interaction, robots' responses to human actions are often pre-designed according to recognized categories and thus stiff. In this paper, we specify a new Interactive Action Translation (IAT) task which aims to learn end-to-end action interaction from unlabeled interactive pairs, removing explicit action recognition. To enable learning on small-scale data, we propose a Paired-Embedding (PE) method for effective and reliable data augmentation. Specifically, our method first utilizes paired relationships to cluster individual actions in an embedding space. Then two actions originally paired can be replaced with other actions in their respective neighborhood, assembling into new pairs. An Act2Act network based on conditional GAN follows to learn from augmented data. Besides, IAT-test and IAT-train scores are specifically proposed for evaluating methods on our task. Experimental results on two datasets show impressive effects and broad application prospects of our method.

cs.CV↗

arXiv2020

Attention-Oriented Action Recognition for Real-Time Human-Robot Interaction

Despite the notable progress made in action recognition tasks, not much work has been done in action recognition specifically for human-robot interaction. In this paper, we deeply explore the characteristics of the action recognition task in interaction scenarios and propose an attention-oriented multi-level network framework to meet the need for real-time interaction. Specifically, a Pre-Attention network is employed to roughly focus on the interactor in the scene at low resolution firstly and then perform fine-grained pose estimation at high resolution. The other compact CNN receives the extracted skeleton sequence as input for action recognition, utilizing attention-like mechanisms to capture local spatial-temporal patterns and global semantic information effectively. To evaluate our approach, we construct a new action dataset specially for the recognition task in interaction scenarios. Experimental results on our dataset and high efficiency (112 fps at 640 x 480 RGBD) on the mobile computing platform (Nvidia Jetson AGX Xavier) demonstrate excellent applicability of our method on action recognition in real-time human-robot interaction.

cs.CV↗

arXiv2020

KGClean: An Embedding Powered Knowledge Graph Cleaning Framework

The quality assurance of the knowledge graph is a prerequisite for various knowledge-driven applications. We propose KGClean, a novel cleaning framework powered by knowledge graph embedding, to detect and repair the heterogeneous dirty data. In contrast to previous approaches that either focus on filling missing data or clean errors violated limited rules, KGClean enables (i) cleaning both missing data and other erroneous values, and (ii) mining potential rules automatically, which expands the coverage of error detecting. KGClean first learns data representations by TransGAT, an effective knowledge graph embedding model, which gathers the neighborhood information of each data and incorporates the interactions among data for casting data to continuous vector spaces with rich semantics. KGClean integrates an active learning-based classification model, which identifies errors with a small seed of labels. KGClean utilizes an efficient PRO-repair strategy to repair errors using a novel concept of propagation power. Extensive experiments on four typical knowledge graphs demonstrate the effectiveness of KGClean in practice.

cs.DB↗

arXiv2020

Self-Guided Adaptation: Progressive Representation Alignment for Domain Adaptive Object Detection

Unsupervised domain adaptation (UDA) has achieved unprecedented success in improving the cross-domain robustness of object detection models. However, existing UDA methods largely ignore the instantaneous data distribution during model learning, which could deteriorate the feature representation given large domain shift. In this work, we propose a Self-Guided Adaptation (SGA) model, target at aligning feature representation and transferring object detection models across domains while considering the instantaneous alignment difficulty. The core of SGA is to calculate "hardness" factors for sample pairs indicating domain distance in a kernel space. With the hardness factor, the proposed SGA adaptively indicates the importance of samples and assigns them different constrains. Indicated by hardness factors, Self-Guided Progressive Sampling (SPS) is implemented in an "easy-to-hard" way during model adaptation. Using multi-stage convolutional features, SGA is further aggregated to fully align hierarchical representations of detection models. Extensive experiments on commonly used benchmarks show that SGA improves the state-of-the-art methods with significant margins, while demonstrating the effectiveness on large domain shift.

cs.CV↗

arXiv2020

Distinguished regular supercuspidal representations

Based on recent work of Kaletha, we apply Hakim--Murnaghan's result to study distinguished regular supercuspidal representations of tamely ramified reductive $p$-adic groups. Assuming $p$ is sufficiently large, we obtain a necessary and sufficient condition for regular supercuspidal representations to be distinguished. We also investigate the relation between the distinction problem and the Langlands functoriality, and confirm a conjecture of Lapid for regular depth-zero or epipelagic supercuspidal representations.

math.RT↗

arXiv2020

Theta lifts and distinction for regular supercuspidal representations

This article has a twofold purpose. First, by recent works of Kaletha and Loke-Ma, we give an explicit description of the local theta correspondence between regular supercuspidal representations in the equal rank symplectic-orthogonal case. Second, based on this description, we show that the local theta correspondence preserves distinction with respect to unramified Galois involutions.

math.RT↗

arXiv2018

Interference-Free Transceiver Design and Signal Detection for Ambient Backscatter Communication Systems over Frequency-Selective Channels

In this letter, we study the ambient backscatter communication systems over frequency-selective channels. Specifically, we propose an interference-free transceiver design to facilitate signal detection at the reader. Our design utilizes the cyclic prefix (CP) of orthogonal frequency-division multiplexing (OFDM) source symbols, which can cancel the signal interference and thus enhance the detection accuracy at the reader. Meanwhile, our design results in no interference on the existing OFDM communication systems. We also suggest a maximum likelihood (ML) detector for the reader and derive two detection thresholds. Simulations are then provided to corroborate our proposed studies.

cs.IT↗

arXiv2018

Transceiver Design for Ambient Backscatter Communication over Frequency-Selective Channels

Existing studies about ambient backscatter communication mostly assume flat-fading channels. However, frequency-selective channels widely exist in many practical scenarios. Therefore, this paper investigates ambient backscatter communication systems over frequency-selective channels. In particular, we propose an interference-free transceiver design to facilitate signal detection at the reader. Our design utilizes the cyclic prefix (CP) of orthogonal frequency-division multiplexing (OFDM) source symbols, which can cancel the signal interference and thus enhance the detection accuracy at the reader. Meanwhile, our design leads to no interference on the existing OFDM communication systems. Next we suggest a chi-square based detector for the reader and derive the optimal detection threshold. Simulations are then provided to corroborate our proposed studies.

cs.IT↗

arXiv2018

A Cost-Sensitive Deep Belief Network for Imbalanced Classification

Imbalanced data with a skewed class distribution are common in many real-world applications. Deep Belief Network (DBN) is a machine learning technique that is effective in classification tasks. However, conventional DBN does not work well for imbalanced data classification because it assumes equal costs for each class. To deal with this problem, cost-sensitive approaches assign different misclassification costs for different classes without disrupting the true data sample distributions. However, due to lack of prior knowledge, the misclassification costs are usually unknown and hard to choose in practice. Moreover, it has not been well studied as to how cost-sensitive learning could improve DBN performance on imbalanced data problems. This paper proposes an evolutionary cost-sensitive deep belief network (ECS-DBN) for imbalanced classification. ECS-DBN uses adaptive differential evolution to optimize the misclassification costs based on training data, that presents an effective approach to incorporating the evaluation measure (i.e. G-mean) into the objective function. We first optimize the misclassification costs, then apply them to deep belief network. Adaptive differential evolution optimization is implemented as the optimization algorithm that automatically updates its corresponding parameters without the need of prior domain knowledge. The experiments have shown that the proposed approach consistently outperforms the state-of-the-art on both benchmark datasets and real-world dataset for fault diagnosis in tool condition monitoring.

cs.LG↗

arXiv2018

A Multi-State Diagnosis and Prognosis Framework with Feature Learning for Tool Condition Monitoring

In this paper, a multi-state diagnosis and prognosis (MDP) framework is proposed for tool condition monitoring via a deep belief network based multi-state approach (DBNMS). For fault diagnosis, a cost-sensitive deep belief network (namely ECS-DBN) is applied to deal with the imbalanced data problem for tool state estimation. An appropriate prognostic degradation model is then applied for tool wear estimation based on the different tool states. The proposed framework has the advantage of automatic feature representation learning and shows better performance in accuracy and robustness. The effectiveness of the proposed DBNMS is validated using a real-world dataset obtained from the gun drilling process. This dataset contains a large amount of measured signals involving different tool geometries under various operating conditions. The DBNMS is examined for both the tool state estimation and tool wear estimation tasks. In the experimental studies, the prediction results are evaluated and compared with popular machine learning approaches, which show the superior performance of the proposed DBNMS approach.

eess.SP↗

arXiv2017

Efficient Column Generation for Cell Detection and Segmentation

We study the problem of instance segmentation in biological images with crowded and compact cells. We formulate this task as an integer program where variables correspond to cells and constraints enforce that cells do not overlap. To solve this integer program, we propose a column generation formulation where the pricing program is solved via exact optimization of very small scale integer programs. Column generation is tightened using odd set inequalities which fit elegantly into pricing problem optimization. Our column generation approach achieves fast stable anytime inference for our instance segmentation problems. We demonstrate on three distinct light microscopy datasets, with several hundred cells each, that our proposed algorithm rapidly achieves or exceeds state of the art accuracy.

cs.CV↗

arXiv2017

Multi-Person Pose Estimation via Column Generation

We study the problem of multi-person pose estimation in natural images. A pose estimate describes the spatial position and identity (head, foot, knee, etc.) of every non-occluded body part of a person. Pose estimation is difficult due to issues such as deformation and variation in body configurations and occlusion of parts, while multi-person settings add complications such as an unknown number of people, with unknown appearance and possible interactions in their poses and part locations. We give a novel integer program formulation of the multi-person pose estimation problem, in which variables correspond to assignments of parts in the image to poses in a two-tier, hierarchical way. This enables us to develop an efficient custom optimization procedure based on column generation, where columns are produced by exact optimization of very small scale integer programs. We demonstrate improved accuracy and speed for our method on the MPII multi-person pose estimation benchmark.

cs.CV↗

arXiv2017

The Candidate Multi-Cut for Cell Segmentation

Two successful approaches for the segmentation of biomedical images are (1) the selection of segment candidates from a merge-tree, and (2) the clustering of small superpixels by solving a Multi-Cut problem. In this paper, we introduce a model that unifies both approaches. Our model, the Candidate Multi-Cut (CMC), allows joint selection and clustering of segment candidates from a merge-tree. This way, we overcome the respective limitations of the individual methods: (1) the space of possible segmentations is not constrained to candidates of a merge-tree, and (2) the decision for clustering can be made on candidates larger than superpixels, using features over larger contexts. We solve the optimization problem of selecting and clustering of candidates using an integer linear program. On datasets of 2D light microscopy of cell populations and 3D electron microscopy of neurons, we show that our method generalizes well and generates more accurate segmentations than merge-tree or Multi-Cut methods alone.

cs.CV↗

arXiv2017

Double Sparsity Kernel Learning with Automatic Variable Selection and Data Extraction

Learning with Reproducing Kernel Hilbert Spaces (RKHS) has been widely used in many scientific disciplines. Because a RKHS can be very flexible, it is common to impose a regularization term in the optimization to prevent overfitting. Standard RKHS learning employs the squared norm penalty of the learning function. Despite its success, many challenges remain. In particular, one cannot directly use the squared norm penalty for variable selection or data extraction. Therefore, when there exists noise predictors, or the underlying function has a sparse representation in the dual space, the performance of standard RKHS learning can be suboptimal. In the literature,work has been proposed on how to perform variable selection in RKHS learning, and a data sparsity constraint was considered for data extraction. However, how to learn in a RKHS with both variable selection and data extraction simultaneously remains unclear. In this paper, we propose a unified RKHS learning method, namely, DOuble Sparsity Kernel (DOSK) learning, to overcome this challenge. An efficient algorithm is provided to solve the corresponding optimization problem. We prove that under certain conditions, our new method can asymptotically achieve variable selection consistency. Simulated and real data results demonstrate that DOSK is highly competitive among existing approaches for RKHS learning.

stat.ME↗

arXiv2017

Distinction of regular depth-zero supercuspidal L-packets

In this paper, we study the relative Langlands program formulated by Dipendra Prasad for Galois symmetric spaces. Under certain assumptions, we confirm the necessary conditions in Prasad's conjecture for regular depth-zero supercuspidal L-packets.

math.RT↗