arXiv subjects

He Zhu

Publications and source records attributed to He Zhu.

102 records · Page 6Linked to original sources

arXiv2021

Graph Collaborative Reasoning

Graphs can represent relational information among entities and graph structures are widely used in many intelligent tasks such as search, recommendation, and question answering. However, most of the graph-structured data in practice suffers from incompleteness, and thus link prediction becomes an important research problem. Though many models are proposed for link prediction, the following two problems are still less explored: (1) Most methods model each link independently without making use of the rich information from relevant links, and (2) existing models are mostly designed based on associative learning and do not take reasoning into consideration. With these concerns, in this paper, we propose Graph Collaborative Reasoning (GCR), which can use the neighbor link information for relational reasoning on graphs from logical reasoning perspectives. We provide a simple approach to translate a graph structure into logical expressions, so that the link prediction task can be converted into a neural logic reasoning problem. We apply logical constrained neural modules to build the network architecture according to the logical expression and use back propagation to efficiently learn the model parameters, which bridges differentiable learning and symbolic reasoning in a unified architecture. To show the effectiveness of our work, we conduct experiments on graph-related tasks such as link prediction and recommendation based on commonly used benchmark datasets, and our graph collaborative reasoning approach achieves state-of-the-art performance.

cs.IR↗

arXiv2021

Exploiting Shared Knowledge from Non-COVID Lesions for Annotation-Efficient COVID-19 CT Lung Infection Segmentation

The novel Coronavirus disease (COVID-19) is a highly contagious virus and has spread all over the world, posing an extremely serious threat to all countries. Automatic lung infection segmentation from computed tomography (CT) plays an important role in the quantitative analysis of COVID-19. However, the major challenge lies in the inadequacy of annotated COVID-19 datasets. Currently, there are several public non-COVID lung lesion segmentation datasets, providing the potential for generalizing useful information to the related COVID-19 segmentation task. In this paper, we propose a novel relation-driven collaborative learning model to exploit shared knowledge from non-COVID lesions for annotation-efficient COVID-19 CT lung infection segmentation. The model consists of a general encoder to capture general lung lesion features based on multiple non-COVID lesions, and a target encoder to focus on task-specific features based on COVID-19 infections. Features extracted from the two parallel encoders are concatenated for the subsequent decoder part. We develop a collaborative learning scheme to regularize feature-level relation consistency of given input and encourage the model to learn more general and discriminative representation of COVID-19 infections. Extensive experiments demonstrate that trained with limited COVID-19 data, exploiting shared knowledge from non-COVID lesions can further improve state-of-the-art performance with up to 3.0% in dice similarity coefficient and 4.2% in normalized surface dice. Our proposed method promotes new insights into annotation-efficient deep learning for COVID-19 infection segmentation and illustrates strong potential for real-world applications in the global fight against COVID-19 in the absence of sufficient high-quality annotations.

eess.IV↗

arXiv2021

Intra-Class Uncertainty Loss Function for Classification

Most classification models can be considered as the process of matching templates. However, when intra-class uncertainty/variability is not considered, especially for datasets containing unbalanced classes, this may lead to classification errors. To address this issue, we propose a loss function with intra-class uncertainty following Gaussian distribution. Specifically, in our framework, the features extracted by deep networks of each class are characterized by independent Gaussian distribution. The parameters of distribution are learned with a likelihood regularization along with other network parameters. The means of the Gaussian play a similar role as the center anchor in existing methods, and the variance describes the uncertainty of different classes. In addition, similar to the inter-class margin in traditional loss functions, we introduce a margin to intra-class uncertainty to make each cluster more compact and reduce the imbalance of feature distribution from different categories. Based on MNIST, CIFAR, ImageNet, and Long-tailed CIFAR analyses, the proposed approach shows improved classification performance, through learning a better class representation.

cs.CV↗

arXiv2020

FakeSafe: Human Level Data Protection by Disinformation Mapping using Cycle-consistent Adversarial Network

The concept of disinformation is to use fake messages to confuse people in order to protect the real information. This strategy can be adapted into data science to protect valuable private and sensitive data. Huge amount of private data are being generated from personal devices such as smart phone and wearable in recent years. Being able to utilize these personal data will bring big opportunities to design personalized products, conduct precision healthcare and many other tasks that were impossible in the past. However, due to privacy, safety and regulation reasons, it is often difficult to transfer or store data in its original form while keeping them safe. Building a secure data transfer and storage infrastructure to preserving privacy is costly in most cases and there is always a concern of data security due to human errors. In this study, we propose a method, named FakeSafe, to provide human level data protection using generative adversarial network with cycle consistency and conducted experiments using both benchmark and real world data sets to illustrate potential applications of FakeSafe.

cs.AI↗

arXiv2020

Patient similarity: methods and applications

Patient similarity analysis is important in health care applications. It takes patient information such as their electronic medical records and genetic data as input and computes the pairwise similarity between patients. Procedures of typical a patient similarity study can be divided into several steps including data integration, similarity measurement, and neighborhood identification. And according to an analysis of patient similarity, doctors can easily find the most suitable treatments. There are many methods to analyze the similarity such as cluster analysis. And during machine learning become more and more popular, Using neural networks such as CNN is a new hot topic. This review summarizes representative methods used in each step and discusses applications of patient similarity networks especially in the context of precision medicine.

cs.LG↗

arXiv2020

Federated machine learning with Anonymous Random Hybridization (FeARH) on medical records

Sometimes electrical medical records are restricted and difficult to centralize for machine learning, which could only be trained in distributed manner that involved many institutions in the process. However, sometimes some institutions are likely to figure out the private data used for training certain models based on the parameters they obtained, which is a violation of privacy and certain regulations. Under those circumstances, we develop an algorithm, called 'federated machine learning with anonymous random hybridization'(abbreviated as 'FeARH'), using mainly hybridization algorithm to eliminate connections between medical record data and models' parameters, which avoid untrustworthy institutions from stealing patients' private medical records. Based on our experiment, our new algorithm has similar AUCROC and AUCPR result compared with machine learning in centralized manner and original federated machine learning, at the same time, our algorithm can greatly reduce data transfer size in comparison with original federated machine learning.

cs.CY↗

arXiv2020

ART: Abstraction Refinement-Guided Training for Provably Correct Neural Networks

Artificial Neural Networks (ANNs) have demonstrated remarkable utility in various challenging machine learning applications. While formally verified properties of their behaviors are highly desired, they have proven notoriously difficult to derive and enforce. Existing approaches typically formulate this problem as a post facto analysis process. In this paper, we present a novel learning framework that ensures such formal guarantees are enforced by construction. Our technique enables training provably correct networks with respect to a broad class of safety properties, a capability that goes well-beyond existing approaches, without compromising much accuracy. Our key insight is that we can integrate an optimization-based abstraction refinement loop into the learning process and operate over dynamically constructed partitions of the input space that considers accuracy and safety objectives synergistically. The refinement procedure iteratively splits the input space from which training data is drawn, guided by the efficacy with which such partitions enable safety verification. We have implemented our approach in a tool (ART) and applied it to enforce general safety properties on unmanned aviator collision avoidance system ACAS Xu dataset and the Collision Detection dataset. Importantly, we empirically demonstrate that realizing safety does not come at the price of much accuracy. Our methodology demonstrates that an abstraction refinement methodology provides a meaningful pathway for building both accurate and correct machine learning networks.

cs.LG↗

arXiv2020

Robustness to Adversarial Attacks in Learning-Enabled Controllers

Learning-enabled controllers used in cyber-physical systems (CPS) are known to be susceptible to adversarial attacks. Such attacks manifest as perturbations to the states generated by the controller's environment in response to its actions. We consider state perturbations that encompass a wide variety of adversarial attacks and describe an attack scheme for discovering adversarial states. To be useful, these attacks need to be natural, yielding states in which the controller can be reasonably expected to generate a meaningful response. We consider shield-based defenses as a means to improve controller robustness in the face of such perturbations. Our defense strategy allows us to treat the controller and environment as black-boxes with unknown dynamics. We provide a two-stage approach to construct this defense and show its effectiveness through a range of experiments on realistic continuous control domains such as the navigation control-loop of an F16 aircraft and the motion control system of humanoid robots.

cs.LG↗

arXiv2019

An Inductive Synthesis Framework for Verifiable Reinforcement Learning

Despite the tremendous advances that have been made in the last decade on developing useful machine-learning applications, their wider adoption has been hindered by the lack of strong assurance guarantees that can be made about their behavior. In this paper, we consider how formal verification techniques developed for traditional software systems can be repurposed for verification of reinforcement learning-enabled ones, a particularly important class of machine learning systems. Rather than enforcing safety by examining and altering the structure of a complex neural network implementation, our technique uses blackbox methods to synthesizes deterministic programs, simpler, more interpretable, approximations of the network that can nonetheless guarantee desired safety properties are preserved, even when the network is deployed in unanticipated or previously unobserved environments. Our methodology frames the problem of neural network verification in terms of a counterexample and syntax-guided inductive synthesis procedure over these programs. The synthesis procedure searches for both a deterministic program and an inductive invariant over an infinite state transition system that represents a specification of an application's control logic. Additional specifications defining environment-based constraints can also be provided to further refine the search space. Synthesized programs deployed in conjunction with a neural network implementation dynamically enforce safety conditions by monitoring and preventing potentially unsafe actions proposed by neural policies. Experimental results over a wide range of cyber-physical applications demonstrate that software-inspired formal verification techniques can be used to realize trustworthy reinforcement learning systems with low overhead.

cs.LG↗

arXiv2018

Distributed Data Vending on Blockchain

Recent advances in blockchain technologies have provided exciting opportunities for decentralized applications. Specifically, blockchain-based smart contracts enable credible transactions without authorized third parties. The attractive properties of smart contracts facilitate distributed data vending, allowing for proprietary data to be securely exchanged on a blockchain. Distributed data vending can transform domains such as healthcare by encouraging data distribution from owners and enabling large-scale data aggregation. However, one key challenge in distributed data vending is the trade-off dilemma between the effectiveness of data retrieval, and the leakage risk from indexing the data. In this paper, we propose a framework for distributed data vending through a combination of data embedding and similarity learning. We illustrate our framework through a practical scenario of distributing and aggregating electronic medical records on a blockchain. Extensive empirical results demonstrate the effectiveness of our framework.

cs.CR↗

arXiv2018

EdgeChain: Blockchain-based Multi-vendor Mobile Edge Application Placement

The state-of-the-art mobile edge applications are generating intense traffic and posing rigorous latency requirements to service providers. While resource sharing across multiple service providers can be a way to maximize the utilization of limited resources at the network edge, it requires a centralized repository maintained by all parties for service providers to share status. Moreover, service providers have to trust each other for resource allocation fairness, which is difficult because of potential conflicts of interest. We propose EdgeChain, a blockchain-based architecture to make mobile edge application placement decisions for multiple service providers. We first formulate a stochastic programming problem minimizing the placement cost for mobile edge application placement scenarios. Based on our model, we present a heuristic mobile edge application placement algorithm. As a decentralized public ledger, the blockchain then takes the logic of our algorithm as the smart contract, with the consideration of resources from all mobile edge hosts participating in the system. The algorithm is agreed by all parties and the results will only be accepted by majority of the mining nodes on the blockchain. When a placement decision is made, an edge host meeting the consumer's latency and budget requirements will be selected at the lowest cost. All placement transactions are stored on the blockchain and are traceable by every mobile edge service provider and application vendor who consumes resources at the mobile edge.

cs.NI↗

arXiv2016

A New Framework for Ranking Vulnerabilities in the Clouds

Qualifying and ranking threat degrees of vulnerabilities in cloud service are known to be full of challenges. Although there have been several efforts aiming to address this problem, most of them are too simple or cannot be applied into cloud infrastructure. This paper aims to propose a novel framework to qualify and rank the vulnerabilities based on their threat degrees in cloud service. Through inputting or constructing service dependency graph, our framework is able to generate the importance degree of each service and the ranking list of all the vulnerabilities in cloud service. Moreover, our framework can be adopted not only into various cloud infrastructures, but also different categories of algorithms according to concrete requirements. To evaluate our framework, we adopt AssetRank algorithm into the framework, and present the whole design of our work. Comprehensive experiments prove the effectiveness of our framework on qualifying and ranking vulnerabilities in cloud service.

cs.CR↗