教学课程资源库 更多>>

关键词： Computer science   Systematic biology   Oncology   Bioinformatics   Artificial intelligence   Datasets   Mutation   Chromosomes   Kinases  

关键词： Computer science   Computer engineering   Artificial intelligence  

摘要： Light field (LF) imaging has gained significant attention due to its recent success in microscopy, 3-dimensional (3D) displaying and rendering, augmented and virtual reality usage. Postprocessing of LF enables us to extract more information from a scene compared to traditional cameras. However, the use of LF is still a research novelty because of the current limitations in capturing high-resolution LF in all of its four dimensions. While researchers are actively improving methods of capturing high-resolution LF's, using simulation, it is possible to explore a high-quality captured LF's properties. The immediate concerns following the LF capture are its storage and processing time. A rich LF occupies a large chunk of memory ---order of multiple gigabytes per LF---. Also, most feature extraction techniques associated with LF postprocessing involve multi-dimensional integration that requires access to the whole LF and is usually time-consuming. Recent advancements in computer processing units made it possible to simulate realistic images using physical-based rendering software. In this work, at first, a transformation function is proposed for building a camera array (CA) to capture the same portion of LF from a scene that a standard plenoptic camera (SPC) can acquire. Using this transformation, LF simulation with similar properties as a plenoptic camera will become trivial in any rendering software. Artificial intelligence (AI) and machine learning (ML) algorithms ---when deployed on the new generation of GPUs--- are faster than ever. It is possible to generate and train large networks with millions of trainable parameters to learn very complex features. Here, residual convolutional neural network (RCNN) structures are employed to build complex networks for compression and feature extraction from an LF. By combining state-of-the-art image compression and RCNN, I have created a compression pipeline. The proposed pipeline's bit per pixel (bpp) ratio is 0.0047 on average.

关键词： Mechanical engineering   Computer science   Operating systems   Unmanned aerial vehicles   Open source software   Visualization   Virtual reality   Sensors  

摘要： Unmanned Aerial Vehicles (UAVs) are becoming more and more integrated into our lives as they are able to fulfill new tasks and become more affordable. Human-UAV Interaction (HUI) therefore is an important topic researcher are working on. As a type of UAVs, quadcopters are characterized with uncertainties in their flying behavior and considered dangerous for humans because of their spinning propellers. Interaction through touch is therefore rarely used as interaction method. Hence, in this work Virtual Reality (VR) is used to simulate a quadcopter to create a platform for virtual HUI. In the development process, Unity was chosen to handle the Virtual Reality application and Gazebo running on the Robot Operating System was used to simulate the quadcopter. The user interface is provided by the Oculus Quest 2 VR headset. Finally, two conducted experiments validated that the system is capable of computing and performing new trajectories of the UAV from physical user inputs. Moreover, the performance of the system was analyzed, and delays of dedicated system components were computed.

关键词： Artificial intelligence   Robotics   Computer science   Automotive engineering  

摘要： Self-Driving cars are a long-lasting ambition for many AI scientists and engineers. In the last decade alone, many self-driving cars like Google Waymo, Tesla Autopilot, Uber, etc. have been roaming the streets of many cities. As a rapidly expanding field, researchers all over the world are attempting to develop more safe and efficient AI agents that can navigate through our cities. However, driving is a very complex task to master even for a human, let alone the challenges in developing robots to do the same. It requires attention and inputs from the surroundings of the car, and it is nearly impossible for us to program all the possible factors affecting this complex task. As a solution, imitation learning was introduced, wherein the agents learn a policy, mapping the observations to the actions through demonstrations given by humans. Through imitation learning, one could easily teach self-driving cars the expected behavior in many scenarios. Despite their autonomous nature, it is undeniable that humans play a vital role in the development and execution of safe and trustworthy self-driving cars and hence form the strongest link in this application of Human-Robot Interaction. Several approaches were taken to incorporate this link between humans and self-driving cars, one of which involves the communication of human's navigational instruction to self-driving cars. The communicative channel provides humans with control over the agent’s decisions as well as the ability to guide them in real-time. In this work, the abilities of imitation learning in creating a self-driving agent that can follow natural language instructions given by humans based on environmental objects’ descriptions were explored. The proposed model architecture is capable of handling latent temporal context in these instructions thus making the agent capable of taking multiple decisions along its course. The work shows promising results that push the boundaries of natural language instructions and thei

关键词： MicroRNAs   Software   Genomics   Biochemistry   Chemistry   Computer science   Genetics  

摘要： Understanding the link between genotype and phenotype is a key question in biological research. In other words, how do changes in DNA sequence give rise to the enormous diversity of life on Earth? In this work, we address that question by focusing on the genomics of convergent phenotypes, or phenotypes that have evolved independently in unrelated species. These natural biological replicates of phenotype evolution allow us to identify genomic regions, either regulatory or protein coding, that have experienced convergent evolution in concordance with convergent evolution of phenotypes, thus linking genomic regions to phenotypes. In this work, we calculate evolutionary rates throughout the mammalian phylogeny for numerous genomic sequences to find concordance between species phenotype and evolutionary rate of sequences to link genomic regions to phenotypes. Chapter one describes three methods associated with quantifying the connection between genomic region evolution and phenotype evolution. First, RERconverge connects genomic regions to phenotypes in a linear regression-based framework. Second, permulations are a statistical extension to RERconverge that allow for rigorous calculation of confidence in associations from RERconverge. Third, proper implementation of branch-site models for convergent positive selection allows for identification of genes potentially driving convergent evolution. Chapter two describes implementation of methods from chapter one to longevity phenotypes in 61 mammal species. We found increased evolutionary constraint in cancer control genes in large, long-lived species, thus likely conferring additional protection from cancer. Species exceptionally long-lived given their size showed increased evolutionary constraint on DNA repair pathways, indicating that efficient DNA repair is important to evolution of extreme lifespan independent of body size. This work provided insight into pan-mammalian genomic mechanisms underlying lifespan. Chapter thre

关键词： Artificial intelligence   Computer science   Information technology   Phase transitions   Job shops   Algorithms   Heuristic  

摘要： Heuristic search algorithms are widely used in both AI planning and the decoding of sequences from deep neural networks. In recent years, several lines of work have highlighted different factors that impact the empirical performance of heuristic search algorithms. However, a principled empirical understanding of the search behavior of these heuristic search algorithms has yet to be developed. Empirical models, such as the phase transition and the heavy-tailed behavior, have been central to the development of empirical understanding of combinatorial search problems such as constraint satisfaction problems (CSP) and satisfiability (SAT). In this dissertation, we investigate the use of empirical models to explain the behavior of heuristic search algorithms in AI planning and neural sequence decoding and support the development of more efficient search algorithms. In AI planning, we develop empirical models for problem difficulty of greedy best first (GBFS), the most commonly used algorithm for satisficing planning. First, we establish the existence of a phase transition in the solubility of planning problems and investigate its implications to problem difficulty. Then, we demonstrate the heavy-tailed behavior of GBFS and provide a deeper understanding of the connection between constrainedness and local minima. Informed by our analysis, we develop a novel variant of GBFS that outperforms the baseline. In neural sequence decoding, we develop empirical models for the performance of beam search, the ubiquitous algorithm for decoding deep sequence models. First, we investigate the empirical problem of performance degradation in beam search. We present an explanatory model based on search discrepancies that generalizes previous observations on the behavior of beam search. Building on our analysis, we present two heuristic techniques that eliminate the problem. Next, we study goal-oriented sequence decoding and show that, similar to GBFS, we observe heavy-tailed behavior. We

关键词： Electrical engineering   Computer science   Systems science   Information technology  

摘要： Advances in communication technologies and computational power have determined a technological shift in the data paradigm. The resulting architecture requires sensors to send local data to the cloud for global processing such as estimation, control, decision and learning, leading to both performance improvement and privacy concerns. This thesis explores the emerging field of private control for Internet of Things, where it bridges dynamical systems and computations on encrypted data, using applied cryptography and information-theoretic tools. Our research contributions are privacy-preserving interactive protocols for cloud-outsourced decisions and data processing, as well as for aggregation over networks in multi-agent systems, both of which are essential in control theory and machine learning. In these settings, we guarantee privacy of the data providers' local inputs over multiple time steps, as well as privacy of the cloud service provider's proprietary information. Specifically, we focus on (i) private solutions to cloud-based constrained quadratic optimization problems from distributed private data; (ii) oblivious distributed weighted sum aggregation; (iii) linear and nonlinear cloud-based control on encrypted data; (iv) private evaluation of cloud-outsourced data-driven control policies with sparsity and low-complexity requirements. In these scenarios, we require computational privacy and stipulate that each participant is allowed to learn nothing more than its own result of the computation. Our protocols employ homomorphic encryption schemes and secure multi-party computation tools with the purpose of performing computations directly on encrypted data, such that leakage of private information at the computing entity is minimized. To this end, we co-design solutions with respect to both control performance and privacy specifications, and we streamline their implementation by exploiting the rich structure of the underlying private data.

关键词： Computer science   Information Technology   Artificial intelligence  

摘要： Convolutional neural networks (CNNs) can be trained with machine learning techniques by using large datasets of images to solve a multitude of useful computer vision tasks. However, CNNs also suffer from a set of vulnerabilities that allow maliciously crafted inputs to affect both their inference and training. A central premise of this dissertation is that these vulnerabilities exhibit a duality when it comes to security and privacy. On the one hand, when computer vision models are applied in safety-critical settings such as autonomous driving, it is important to identify failures that can be exploited by malicious parties early on so that system designers can plan for novel threat models. On the other hand, when machine learning models themselves are being used in a malicious or unauthorized manner, such vulnerabilities can be leveraged to protect data creators from harmful effects of these models (such as privacy degradation) and enforce finer-grained “access” controls over the data. This work studies security and privacy issues in three scenarios where machine learning is applied for visual tasks. The first contribution of this work is to identify a vulnerability in models that are likely to be deployed to identify road signs in autonomous vehicles. It demonstrates that an attacker with no digital access to a self-driving car’s computers can nevertheless cause dangerous behavior by modifying the appearance of physical objects. Next, this dissertation considers scenarios where machine learning models are applied in a way that degrades individual privacy. The dissertation proposes a scheme -- nicknamed FoggySight -- in which a community of users volunteer adversarial modified photos (“decoys”) that poison the facial search database and throw off searches in it. Finally, machine learning models may be trained on data without authorization to do so. This dissertation considers scenarios where image owners might wish to share their visual data widely for human consump

关键词： Alternative Energy   Artificial intelligence   Atmospheric sciences   Computer science   Energy   Operations research  

摘要： Complex non-linear prediction systems have become ubiquitous in numerous decision making and other socio-technical systems. In recent years, the increased adoption and use of these complex non-linear systems has been dominated by universal approximators such as neural networks and Gaussian Processes. These systems’ applications span a large number of critical domains, including transportation, drug design, law enforcement, financial services, energy planning, and pandemic forecasting. The aforementioned critical nature of the application domains necessitates the need to study the inference methods for training or calibration of these systems’ parameters. Further to this, inference methods coupled with estimators of the uncertainty around the system’s predictions and measures of the relative influence of its inputs aid in managing the very high societal risks associated with incorrect predictions. This thesis investigates probabilistic parameter inference methods that provide both the required uncertainty and relevance measures. We first introduce Metropolis Hastings (MH) and Hybrid Monte Carlo (HMC) methods for parameter inference in Bayesian Neural Networks (BNNs) with applications in credit risk modelling and South African wind energy resource planning. We further utilise a Separable Shadow Hamiltonian Hybrid Monte Carlo (S2HMC) method for the first time in the inference of BNN parameters. S2HMC addresses traditional MCMC methods’ discretisation constraints by using a perturbed Hamiltonian, which is conserved at a higher-order by the numerical integration scheme. Experimental results on wind energy and credit datasets find that S2HMC yields higher effective sample sizes than the competing Hybrid Monte Carlo (HMC). The predictive performance of S2HMC and HMC based BNNs is found to be similar. We thirdly perform hierarchical inference for BNN parameters by combining the S2HMC sampler with Gibbs sampling of hyperparameters for Automatic Relevance Determination (ARD).

关键词： Servers   Browsers   Web pages   Time-frequency analysis   Prefetching   Prediction algorithms   Computer science   Locally normalized intervals   proxy   web cache   pre-fetching  

摘要： Caching is one of the most promising areas in the field of future internet architecture like Information-centric Networking, Software Defined Networking, and IoT. In Web caching, most of the web content is readily available across the network, even if the webserver is not reachable. Several existing traditional caching methods and cache replacement strategies are evaluated based on the metrics like hit ratio and byte hit Ratio. However, these metrics have not been improved over the period because of the traditional caching policies. So, in this paper, we have used an intelligent function like locally normalized intervals of page visit, website duration, users' interest between user groups is proposed. These intervals are combined with multiple distance metrics like Manhattan, squared Euclidean, and 3-,4-,5-norm Minkowski. In order to obtain significant common user navigation patterns, the clustering relation between the users using different intervals and distances is thoroughly analyzed. These patterns are successfully coupled with greedy web cache replacement strategies to improve the efficiency of the proposed web cache system. Particularly for improving the caching metrics more, we used an AI-based intelligent approach like Random Forest classifier to boost the prefetch buffer performance and achieves the maximum hit rate of 0.89, 0.90, and byte hit rate of 0.87, 0.89 for Greedy Dual Size Frequency and Weighted Greedy Dual Size Frequency algorithms, respectively. Our experiments show good hit/byte hit rates than the frequently used algorithms like least recently used and least frequently used.