arXiv subjects

Shuo Wang

Publications and source records attributed to Shuo Wang.

At least 487 records · Page 27Linked to original sources

arXiv2022

InvNorm: Domain Generalization for Object Detection in Gastrointestinal Endoscopy

Domain Generalization is a challenging topic in computer vision, especially in Gastrointestinal Endoscopy image analysis. Due to several device limitations and ethical reasons, current open-source datasets are typically collected on a limited number of patients using the same brand of sensors. Different brands of devices and individual differences will significantly affect the model's generalizability. Therefore, to address the generalization problem in GI(Gastrointestinal) endoscopy, we propose a multi-domain GI dataset and a light, plug-in block called InvNorm(Invertible Normalization), which could achieve a better generalization performance in any structure. Previous DG(Domain Generalization) methods fail to achieve invertible transformation, which would lead to some misleading augmentation. Moreover, these models would be more likely to lead to medical ethics issues. Our method utilizes normalizing flow to achieve invertible and explainable style normalization to address the problem. The effectiveness of InvNorm is demonstrated on a wide range of tasks, including GI recognition, GI object detection, and natural image recognition.

eess.IV↗

arXiv2022

A New Method of Construction of Permutation Trinomials with Coefficients 1

Permutation polynomials over finite fields are an interesting and constantly active research subject of study for many years. They have important applications in areas of mathematics and engineering. In recent years, permutation binomials and permutation trinomials attract people's interests due to their simple algebraic forms. In this paper, by reversely using Tu's method for the characterization of permutation polynomials with exponents of Niho type, we propose a new method to construct permutation trinomials with coefficients 1. Moreover, we give the explicit compositional inverses of a class of permutation trinomials for a special case.

math.NT↗

arXiv2022

Exchange bias in van der Waals MnBi$_2$Te$_4$/Cr$_2$Ge$_2$Te$_6$ heterostructure

The layered van der Waals (vdW) material MnBi$_2$Te$_4$ is an intrinsic magnetic topological insulator with various topological phases such as quantum anomalous Hall effect (QAHE) and axion states. However, both the zero-field and high-temperature QAHE are not easy to realize. It is theoretically proposed that the exchange bias can be introduced in the MnBi2Te4/ferromagnetic (FM) insulator heterostructures and thus opens the surface states gap, making it easier to realize the zero-field or high-temperature QAHE. Here we report the electrically tunable exchange bias in the van der Waals MnBi$_2$Te$_4$/Cr$_2$Ge$_2$Te$_6$ heterostructure. The exchange bias emerges over a critical magnetic field and reaches the maximum value near the magnetic band gap. Moreover, the exchange bias was experienced by the antiferromagnetic (AFM) MnBi$_2$Te$_4$ layer rather than the FM layer. Such van der Waals heterostructure provides a promising platform to study the novel exchange bias effect and explore the possible high-temperature QAHE.

cond-mat.mtrl-sci↗

arXiv2022

A Roadmap for Big Model

With the rapid development of deep learning, training Big Models (BMs) for multiple downstream tasks becomes a popular paradigm. Researchers have achieved various outcomes in the construction of BMs and the BM application in many fields. At present, there is a lack of research work that sorts out the overall progress of BMs and guides the follow-up research. In this paper, we cover not only the BM technologies themselves but also the prerequisites for BM training and applications with BMs, dividing the BM review into four parts: Resource, Models, Key Technologies and Application. We introduce 16 specific BM-related topics in those four parts, they are Data, Knowledge, Computing System, Parallel Training System, Language Model, Vision Model, Multi-modal Model, Theory&Interpretability, Commonsense Reasoning, Reliability&Security, Governance, Evaluation, Machine Translation, Text Generation, Dialogue and Protein Research. In each topic, we summarize clearly the current studies and propose some future research directions. At the end of this paper, we conclude the further development of BMs in a more general view.

cs.LG↗

arXiv2022

Attention in Attention: Modeling Context Correlation for Efficient Video Classification

Attention mechanisms have significantly boosted the performance of video classification neural networks thanks to the utilization of perspective contexts. However, the current research on video attention generally focuses on adopting a specific aspect of contexts (e.g., channel, spatial/temporal, or global context) to refine the features and neglects their underlying correlation when computing attentions. This leads to incomplete context utilization and hence bears the weakness of limited performance improvement. To tackle the problem, this paper proposes an efficient attention-in-attention (AIA) method for element-wise feature refinement, which investigates the feasibility of inserting the channel context into the spatio-temporal attention learning module, referred to as CinST, and also its reverse variant, referred to as STinC. Specifically, we instantiate the video feature contexts as dynamics aggregated along a specific axis with global average and max pooling operations. The workflow of an AIA module is that the first attention block uses one kind of context information to guide the gating weights calculation of the second attention that targets at the other context. Moreover, all the computational operations in attention units act on the pooled dimension, which results in quite few computational cost increase ($<$0.02\%). To verify our method, we densely integrate it into two classical video network backbones and conduct extensive experiments on several standard video classification benchmarks. The source code of our AIA is available at \url{https://github.com/haoyanbin918/Attention-in-Attention}.

cs.CV↗

arXiv2022

Towards Web Phishing Detection Limitations and Mitigation

Web phishing remains a serious cyber threat responsible for most data breaches. Machine Learning (ML)-based anti-phishing detectors are seen as an effective countermeasure, and are increasingly adopted by web-browsers and software products. However, with an average of 10K phishing links reported per hour to platforms such as PhishTank and VirusTotal (VT), the deficiencies of such ML-based solutions are laid bare. We first explore how phishing sites bypass ML-based detection with a deep dive into 13K phishing pages targeting major brands such as Facebook. Results show successful evasion is caused by: (1) use of benign services to obscure phishing URLs; (2) high similarity between the HTML structures of phishing and benign pages; (3) hiding the ultimate phishing content within Javascript and running such scripts only on the client; (4) looking beyond typical credentials and credit cards for new content such as IDs and documents; (5) hiding phishing content until after human interaction. We attribute the root cause to the dependency of ML-based models on the vertical feature space (webpage content). These solutions rely only on what phishers present within the page itself. Thus, we propose Anti-SubtlePhish, a more resilient model based on logistic regression. The key augmentation is the inclusion of a horizontal feature space, which examines correlation variables between the final render of suspicious pages against what trusted services have recorded (e.g., PageRank). To defeat (1) and (2), we correlate information between WHOIS, PageRank, and page analytics. To combat (3), (4) and (5), we correlate features after rendering the page. Experiments with 100K phishing/benign sites show promising accuracy (98.8%). We also obtained 100% accuracy against 0-day phishing pages that were manually crafted, comparing well to the 0% recorded by VT vendors over the first four days.

cs.CR↗

arXiv2022

CAFE: Learning to Condense Dataset by Aligning Features

Dataset condensation aims at reducing the network training effort through condensing a cumbersome training set into a compact synthetic one. State-of-the-art approaches largely rely on learning the synthetic data by matching the gradients between the real and synthetic data batches. Despite the intuitive motivation and promising results, such gradient-based methods, by nature, easily overfit to a biased set of samples that produce dominant gradients, and thus lack global supervision of data distribution. In this paper, we propose a novel scheme to Condense dataset by Aligning FEatures (CAFE), which explicitly attempts to preserve the real-feature distribution as well as the discriminant power of the resulting synthetic set, lending itself to strong generalization capability to various architectures. At the heart of our approach is an effective strategy to align features from the real and synthetic data across various scales, while accounting for the classification of real samples. Our scheme is further backed up by a novel dynamic bi-level optimization, which adaptively adjusts parameter updates to prevent over-/under-fitting. We validate the proposed CAFE across various datasets, and demonstrate that it generally outperforms the state of the art: on the SVHN dataset, for example, the performance gain is up to 11%. Extensive experiments and analyses verify the effectiveness and necessity of proposed designs.

cs.CV↗

arXiv2022

Integrating Vectorized Lexical Constraints for Neural Machine Translation

Lexically constrained neural machine translation (NMT), which controls the generation of NMT models with pre-specified constraints, is important in many practical scenarios. Due to the representation gap between discrete constraints and continuous vectors in NMT models, most existing works choose to construct synthetic data or modify the decoding algorithm to impose lexical constraints, treating the NMT model as a black box. In this work, we propose to open this black box by directly integrating the constraints into NMT models. Specifically, we vectorize source and target constraints into continuous keys and values, which can be utilized by the attention modules of NMT models. The proposed integration method is based on the assumption that the correspondence between keys and values in attention modules is naturally suitable for modeling constraint pairs. Experimental results show that our method consistently outperforms several representative baselines on four language pairs, demonstrating the superiority of integrating vectorized lexical constraints.

cs.CL↗

arXiv2022

Calibrating DFT formation enthalpy calculations by multi-fidelity machine learning

Machine learning materials properties measured by experiments is valuable yet difficult due to the limited amount of experimental data. In this work, we use a multi-fidelity random forest model to learn the experimental formation enthalpy of materials with prediction accuracy higher than the empirically corrected PBE functional (PBEfe) and meta-GGA functional (SCAN), and it outperforms the hotly studied deep neural-network based representation learning and transfer learning. We then use the model to calibrate the DFT formation enthalpy in the Materials Project database, and discover materials with underestimated stability. The multi-fidelity model is also used as a data-mining approach to find how DFT deviates from experiments by the explaining the model output.

cond-mat.mtrl-sci↗

arXiv2022

MSP: Multi-Stage Prompting for Making Pre-trained Language Models Better Translators

Prompting has recently been shown as a promising approach for applying pre-trained language models to perform downstream tasks. We present Multi-Stage Prompting (MSP), a simple and automatic approach for leveraging pre-trained language models to translation tasks. To better mitigate the discrepancy between pre-training and translation, MSP divides the translation process via pre-trained language models into multiple separate stages: the encoding stage, the re-encoding stage, and the decoding stage. During each stage, we independently apply different continuous prompts for allowing pre-trained language models better shift to translation tasks. We conduct extensive experiments on three translation tasks. Experiments show that our method can significantly improve the translation performance of pre-trained language models.

cs.CL↗

arXiv2022

An Efficient Training Approach for Very Large Scale Face Recognition

Face recognition has achieved significant progress in deep learning era due to the ultra-large-scale and welllabeled datasets. However, training on the outsize datasets is time-consuming and takes up a lot of hardware resource. Therefore, designing an efficient training approach is indispensable. The heavy computational and memory costs mainly result from the million-level dimensionality of thefully connected (FC) layer. To this end, we propose a novel training approach, termed Faster Face Classification (F2C), to alleviate time and cost without sacrificing the performance. This method adopts Dynamic Class Pool (DCP) for storing and updating the identities features dynamically, which could be regarded as a substitute for the FC layer. DCP is efficiently time-saving and cost-saving, as its smaller size with the independence from the whole face identities together. We further validate the proposed F2C method across several face benchmarks and private datasets, and display comparable results, meanwhile the speed is faster than state-of-the-art FC-based methods in terms of recognition accuracy and hardware costs. Moreover, our method is further improved by a well-designed dual data loader including indentity-based and instancebased loaders, which makes it more efficient for the updating DCP parameters.

cs.CV↗

arXiv2022

OCTOPUS: Overcoming Performance andPrivatization Bottlenecks in Distributed Learning

The diversity and quantity of data warehouses, gathering data from distributed devices such as mobile devices, can enhance the success and robustness of machine learning algorithms. Federated learning enables distributed participants to collaboratively learn a commonly-shared model while holding data locally. However, it is also faced with expensive communication and limitations due to the heterogeneity of distributed data sources and lack of access to global data. In this paper, we investigate a practical distributed learning scenario where multiple downstream tasks (e.g., classifiers) could be efficiently learned from dynamically-updated and non-iid distributed data sources while providing local data privatization. We introduce a new distributed/collaborative learning scheme to address communication overhead via latent compression, leveraging global data while providing privatization of local data without additional cost due to encryption or perturbation. This scheme divides learning into (1) informative feature encoding, and transmitting the latent representation of local data to address communication overhead; (2) downstream tasks centralized at the server using the encoded codes gathered from each node to address computing overhead. Besides, a disentanglement strategy is applied to address the privatization of sensitive components of local data. Extensive experiments are conducted on image and speech datasets. The results demonstrate that downstream tasks on the compact latent representations with the privatization of local data can achieve comparable accuracy to centralized learning.

cs.LG↗

arXiv2022

Adversarial Defense by Latent Style Transformations

Machine learning models have demonstrated vulnerability to adversarial attacks, more specifically misclassification of adversarial examples. In this paper, we investigate an attack-agnostic defense against adversarial attacks on high-resolution images by detecting suspicious inputs. The intuition behind our approach is that the essential characteristics of a normal image are generally consistent with non-essential style transformations, e.g., slightly changing the facial expression of human portraits. In contrast, adversarial examples are generally sensitive to such transformations. In our approach to detect adversarial instances, we propose an in\underline{V}ertible \underline{A}utoencoder based on the \underline{S}tyleGAN2 generator via \underline{A}dversarial training (VASA) to inverse images to disentangled latent codes that reveal hierarchical styles. We then build a set of edited copies with non-essential style transformations by performing latent shifting and reconstruction, based on the correspondences between latent codes and style transformations. The classification-based consistency of these edited copies is used to distinguish adversarial instances.

cs.CV↗

arXiv2022

Defeating Misclassification Attacks Against Transfer Learning

Transfer learning is prevalent as a technique to efficiently generate new models (Student models) based on the knowledge transferred from a pre-trained model (Teacher model). However, Teacher models are often publicly available for sharing and reuse, which inevitably introduces vulnerability to trigger severe attacks against transfer learning systems. In this paper, we take a first step towards mitigating one of the most advanced misclassification attacks in transfer learning. We design a distilled differentiator via activation-based network pruning to enervate the attack transferability while retaining accuracy. We adopt an ensemble structure from variant differentiators to improve the defence robustness. To avoid the bloated ensemble size during inference, we propose a two-phase defence, in which inference from the Student model is firstly performed to narrow down the candidate differentiators to be assembled, and later only a small, fixed number of them can be chosen to validate clean or reject adversarial inputs effectively. Our comprehensive evaluations on both large and small image recognition tasks confirm that the Student models with our defence of only 5 differentiators are immune to over 90% of the adversarial inputs with an accuracy loss of less than 10%. Our comparison also demonstrates that our design outperforms prior problematic defences.

cs.LG↗

arXiv2022

GenAD: General Representations of Multivariate Time Seriesfor Anomaly Detection

The reliability of wireless base stations in China Mobile is of vital importance, because the cell phone users are connected to the stations and the behaviors of the stations are directly related to user experience. Although the monitoring of the station behaviors can be realized by anomaly detection on multivariate time series, due to complex correlations and various temporal patterns of multivariate series in large-scale stations, building a general unsupervised anomaly detection model with a higher F1-score remains a challenging task. In this paper, we propose a General representation of multivariate time series for Anomaly Detection(GenAD). First, we pre-train a general model on large-scale wireless base stations with self-supervision, which can be easily transferred to a specific station anomaly detection with a small amount of training data. Second, we employ Multi-Correlation Attention and Time-Series Attention to represent the correlations and temporal patterns of the stations. With the above innovations, GenAD increases F1-score by total 9% on real-world datasets in China Mobile, while the performance does not significantly degrade on public datasets with only 10% of the training data.

cs.NI↗

arXiv2022

Flexible Design on Deterministic IP Networking for Mixed Traffic Transmission

Deterministic IP (DIP) networking is a promising technique that can provide delay-bounded transmission in large-scale networks. Nevertheless, DIP faces several challenges in the mixed traffic scenarios, including (i) the capability of ultra-low latency communications, (ii) the simultaneous satisfaction of diverse QoS requirements, and (iii) the network efficiency. The problems are more formidable in the dynamic surroundings without prior knowledge of traffic demands. To address the above-mentioned issues, this paper designs a flexible DIP (FDIP) network. In the proposed network, we classify the queues at the output port into multiple groups. Each group operates with different cycle lengths. FDIP can assign the time-sensitive flows with different groups, hence delivering diverse QoS requirements, simultaneously. The ultra-low latency communication can be achieved by specific groups with short cycle lengths. Moreover, the flexible scheduling with diverse cycle lengths improves resource utilization, hence increasing the throughput (i.e., the number of acceptable time-sensitive flows). We formulate a throughput maximization problem that jointly considers the admission control, transmission path selection, and cycle length assignment. A branch and bound (BnB)-based heuristic is developed. Simulation results show that the proposed FDIP significantly outperforms the standard DIP in terms of both the throughput and the latency guarantees.

cs.NI↗

arXiv2022

Large-scale Deterministic Transmission among IEEE 802.1Qbv Time-Sensitive Networks

IEEE 802.1Qbv (TAS) is the most widely used technique in Time-Sensitive Networking (TSN) which aims to provide bounded transmission delays and ultra-low jitters in industrial local area networks. With the development of emerging technologies (e.g., cloud computing), many wide-range time-sensitive network services emerge, such as factory automation, connected vehicles, and smart grids. Nevertheless, TAS is a Layer 2 technique for local networks, and cannot provide large-scale deterministic transmission. To tackle this problem, this paper proposes a hierarchical network containing access networks and a core network. Access networks perform TAS to aggregate time-sensitive traffic. In the core network, we exploit DIP (a well-known deterministic networking mechanism for backbone networks) to achieve long-distance deterministic transmission. Due to the differences between TAS and DIP, we design cross-domain transmission mechanisms at the edge of access networks and the core network to achieve seamless deterministic transmission. We also formulate the end-to-end scheduling to maximize the amount of accepted time-sensitive traffic. Experimental simulations show that the proposed network can achieve end-to-end deterministic transmission even in high-loaded scenarios.

cs.NI↗

arXiv2022

Similarity-based Gray-box Adversarial Attack Against Deep Face Recognition

The majority of adversarial attack techniques perform well against deep face recognition when the full knowledge of the system is revealed (\emph{white-box}). However, such techniques act unsuccessfully in the gray-box setting where the face templates are unknown to the attackers. In this work, we propose a similarity-based gray-box adversarial attack (SGADV) technique with a newly developed objective function. SGADV utilizes the dissimilarity score to produce the optimized adversarial example, i.e., similarity-based adversarial attack. This technique applies to both white-box and gray-box attacks against authentication systems that determine genuine or imposter users using the dissimilarity score. To validate the effectiveness of SGADV, we conduct extensive experiments on face datasets of LFW, CelebA, and CelebA-HQ against deep face recognition models of FaceNet and InsightFace in both white-box and gray-box settings. The results suggest that the proposed method significantly outperforms the existing adversarial attack techniques in the gray-box setting. We hence summarize that the similarity-base approaches to develop the adversarial example could satisfactorily cater to the gray-box attack scenarios for de-authentication.

cs.CV↗