教学课程资源库 更多>>

关键词： Bioinformatics   Artificial intelligence   Computer science  

摘要： High-throughput DNA sequencing and related biotechnologies revolutionized our understanding of human genomics and diseases with genetic component, particularly of cancer -- one of the leading causes of death world-wide. Despite the progress in cancer research and availability of over 150 FDA-approved anti-cancer drugs, the cancer treatment is often unsuccessful. Identifying the best cancer treatment using computational models to personalize drug response prediction holds great promise to improve patient’s chances of successful recovery. Unfortunately, the computational task of predicting drug response remains very challenging. In this thesis I develop a deep latent-variable machine learning model with amortized variational inference that improves accuracy of drug response prediction over the currently used models. Besides increased expressiveness of this model thanks to parameterization by neural networks, the achieved improvement stems from integration of drug-induced perturbation profiles, a resource not fully utilized before. Clinical trial datasets of cancer treatments which also include genomic characterization of the tumours are small and scarce, therefore for the vast majority of drugs only responses in pre-clinical biological models are available. To this end, I assess applicability of popular domain adaptation approach, based on domain-invariant representation learning, to the drug response prediction task. I conclude that necessary conditions of successful domain adaptation are often not satisfied in the available datasets and as such many current methods are misguided. Last but not least, in this thesis I also contribute to the area of non-invasive prenatal testing. Using a hidden Markov model I propose a method for analysis of cell-free DNA fragments isolated from maternal plasma that also contain admixture of DNA fragments derived from the fetal genome. Here presented method is a first proof-of-concept for non-invasive sub-chromosomal CNV detection.

关键词： Agriculture   Computer science  

摘要： Precision agriculture technologies are not secure enough to protect the food and agriculture sector from cyber-attack. The reliance on proprietary data and algorithm-driven machinery is increasing, but the knowledge of, or accessibility to cybersecurity methods is not; therefore, it is important to establish policies and laws that underline the importance of securing the food and agriculture sector and to research effective methods of protection. The purpose of this research was to assess the current methods of securing industrial control systems and supervisory control and data acquisition (ICS/SCADA) systems in the United States food and agriculture sector and to determine how the active cyber defense method could improve security in agriculture. This study examined cybersecurity threats to the agriculture industry and the vulnerabilities those threats seek to exploit. The study reviewed the current security measures in smart agriculture and policies and programs that address threats to critical infrastructure. Finally, this study explored the active cyber defense cycle and examined how it could enhance the security of ICS/SCADA systems in smart agriculture. This research found that there is no comprehensive framework that addresses all the cybersecurity needs of precision agriculture systems, but that the active cyber defense cycle offers an excellent baseline for the development of such a framework. More cybersecurity education and further research into comprehensive protections for ICS and SCADA systems need to be made a priority to protect the food and agriculture sector. Keywords: Cybersecurity, Dr. Donnie Wendt, food and agriculture sector, precision agriculture, smart farming, active defense, active cyber defense cycle

关键词： Engineering   Computer science  

摘要： The traditional access control system suffers from the problem of separation of data ownership and management. It poses data security issues in application scenarios such as cloud computing and blockchain where the data owners either do not trust the data storage provider or even do not know who would have access to their data once they are appended to the chain. In these scenarios, the data owner actually loses control of the data once they are uploaded to the outside storage. Encryption-before-uploading is the way to solve this issue, however traditional encryption schemes such as AES, RSA, ECC, bring about great overheads in key management on the data owner end and could not provide fine-grained access control as well. Attribute-Based Encryption (ABE) is a cryptographic way to implement attribute-based access control, which is a fine-grained access control model, thus solving all aforementioned issues. With ABE, the data owner would encrypt the data by a self-defined access control policy before uploading the data. The access control policy is an AND-OR boolean formula over attributes. Only users with attributes that satisfy the access control policy could decrypt the ciphertext. However the existing ABE schemes do not provide some important features in practical applications, e.g., user revocation and attribute expiration. Furthermore, most existing work focus on how to use ABE to protect cloud stored data, while not the blockchain applications. The main objective of this thesis is to provide solutions to add two important features of the ABE schemes, i.e., user revocation and attribute expiration, and also provide a practical trust framework for using ABE to protect blockchain data. To add the feature of user revocation, I propose to add user's hierarchical identity into the private attribute key. In this way, only users whose identity is not revoked and attributes satisfy the access control policy could decrypt the ciphertext. To add the feature of attribute e

关键词： Computer science  

摘要： Convex optimization has many applications to fields as diverse as machine learning, control, finance, and signal and image processing. Using convex optimization in an application requires either developing a custom solver or converting the problem into a standard form. Both of these tasks require expertise, and are time-consuming and error prone. An alternative is to use a domain-specific language (DSL) for convex optimization, which allows the user to specify the problem in a natural way that follows the math; this specification is then automatically converted into the standard form required by generic solvers. In this thesis we demonstrate that DSLs for convex optimization are easy to use, scale to large problems, and can be extended to useful classes of non-convex problems. We begin with a discussion of CVXPY, a widely-used DSL for convex optimization. We present several examples of modeling optimization problems with CVXPY and highlight the novel features and modeling paradigms CVXPY introduced. We next illustrate how DSLs for convex optimization such as CVXPY can be extended to efficiently handle large-scale optimization problems involving structured linear operators. We call our approach matrix-free convex optimization modeling. We conclude with an exploration of non-convex optimization using convex optimization as a black-box method. In particular, we consider approximate minimization of convex functions over non-convex sets via an ADMM-based heuristic that solves a series of convex subproblems. Our approach lends itself to expression as a DSL, which we call NCVX.

关键词： Computer science  

摘要： Runtime monitoring of distributed systems to perform predicate detection is critical as well as a challenging task. It is critical because it ensures the reliability of the system by detecting all possible violations of system requirements. It is challenging because to guarantee lack of violations one has to analyze every possible ordering of system events and this is an expensive task. In this report, we focus on ordering events in a system run using HLC (Hybrid Logical Clocks) timestamps, which are O(1) sized timestamps, and present some efficient algorithms to perform predicate detection using HLC. Since, with HLC, the runtime monitor cannot find all possible orderings of systems events, we present a new type of clock called Biased Hybrid Logical Clocks (BHLC), that are capable of finding more possible orderings than HLC. Thus we show that BHLC based predicate detection can find more violations than HLC based predicate detection. Since predicate detection based on both HLC and BHLC do not guarantee detection of all possible violations in a system run, we present an SMT (Satisfiability Modulo Theories) solver based predicate detection approach, that guarantees the detection of all possible violations in a system run. While a runtime monitor that performs predicate detection using SMT solvers is accurate, the time taken by the solver to detect the presence or absence of a violation can be high. To reduce the time taken by the runtime monitor, we propose the use of an efficient two-layered monitoring approach, where the first layer of the monitor is efficient but less accurate and the second layer is accurate but less efficient. Together they reduce the overall time taken to perform predicate detection drastically and also guarantee detection of all possible violations.

关键词： Computer science  

摘要： Legacy binaries need to continue functioning even when no source code has been preserved, to support the workflows of government and industry. The binaries often lack recent improvements in compiler design and software engineering practices, causing them to be slower and less secure than modern binaries. Binary rewriting seeks to patch, optimize, instrument, or harden binaries to bridge this gap, but existing practice is limited by the underlying static analysis. We created a framework, BinRec, to use dynamic analysis to lift binaries to LLVM IR then recompile them, which overcomes the limitations of static *** protection of software against memory corruption exploits has a rich history, which this thesis both systematizes and extends. We present a study of the performance, precision, and security of control-flow integrity (CFI). Data-only attacks can bypass CFI, and so we present a defense against these attacks. The application of these hardening techniques to binaries deserves further study, and we discuss the extent to which security hardening can be applied to recompiled *** dissertation presents building blocks for the securing of legacy binaries using dynamic analysis, which we hope will become a dominant paradigm in the secure software ecosystem of tomorrow.

关键词： dynamic cellular manufacturing systems   worker assignment   multiple process route   system reconfiguration   part operation trade-off   subcontracting part operation   simulated annealing-based genetic algorithm  

摘要： In this study, an integrated mathematical model for the cell formation problem is proposed considering the dynamic production environment. The proposed model yields, manufacturing cells, part families and worker's assignment simultaneously by allowing a cubic search space of 'machine-part-worker' in the CMS. The resources are aggregated into manufacturing cells based on the optimal process route among the user specified multiple routes. The model interprets flexibility in the processing of subsets of a part operation sequence in the different production mode (internal production/subcontracting part operation). It is a tangible advantage during unavailability of worker and unexpected machine break down occurring in the real world. The proposed cell formation problem has been solved by using a simulated annealing-based genetic algorithm (SAGA). The algorithm imparts synergy effect to improve intensification, diversification in the cubic search space and increases the possibility of achieving near-optimum solutions. To evaluate the computational performance of the proposed approach the algorithm is tested on a number of randomly generated instances. The results substantiate the efficiency of the proposed approach by minimising overall cost. [Received: 17 August 2018;Accepted: 28 July 2019]

ISBN: (纸本)9787111645306

关键词： Mathematics   Transportation   Computer science  

摘要： Road closures due to adverse and severe weather is nothing unusual in Wyoming. One particular stretch of road way, Interstate 80, will be investigated. In this initial data driven study we use Machine Learning techniques to establish a predictive model for road closures in this region using sensor data and road closure data. As the results suggest, such a data driven model will be beneficial in both establishing legitimate road closures and improving general safety in this region during periods of ad verse and severe weather.

关键词： Computer science  

摘要： The purpose of this research is to apply four methods on two data sets, a Synthetic dataset and a Real-World dataset, and compare the results to each other with the intention of arriving at methods to prevent fraud. Methods used include Logistic Regression, Isolation Forest, Ensemble Method and Generative Adversarial Networks. Results show that all four models achieve accuracies between 91% and 99% except Isolation Forest gave 69% accuracy for the Synthetic dataset. The four models detect fraud well when built on a training set and tested with a test set. Logistic Regression achieves good results with less computational eorts. Isolation Forest achieve lower results accuracies when the data is sparse and not preprocessed correctly. Ensemble Models achieve the highest accuracy for both datasets. GAN achieves good results but overts if a big number of epochs was used. Future work could incorporate other classiers.