教学课程资源库 更多>>

关键词： Applied mathematics   Biomedical engineering   Computer science  

摘要： Biomedical time series contain rich information about human systems, however, effective algorithms for analyzing long-term physiological time series have not yet been developed because of the huge volume size, high dimensionality and large noise nature of the data. Motivated by such challenging task, we proposed a novel spectral embedding algorithm, which we coined Robust and Scalable Embedding via Landmark Diffusion (ROSELAND). The solution is a generic and not limited to analyze physiological waveforms. In short, we measure the affinity between two points via a set of landmarks, which is composed of a small number of points, and "diffuse'' on the dataset via the landmark set to achieve a spectral embedding. The algorithm is applied to study the arterial blood pressure waveform dynamics during a liver transplant operation lasting for 12 hours long. In addition, we show that Roseland is not only numerically scalable, but also preserves the geometric properties via its diffusion nature under the manifold setup; that is, we theoretically explore the asymptotical behavior of Roseland under the manifold setup, and provide a L∞ spectral convergence with a rate. Moreover, we offer a high dimensional noise analysis with the help of Gaussian approximation, and show that Roseland is robust to noise.

关键词： Mechanical engineering   Computer science   Applied mathematics   Simulation   Random variables   Estimates   Eigen values   Noise   Literature reviews   Algorithms   Kalman filters   Dynamical systems  

关键词： Robotics   Computer science   Artificial intelligence  

摘要： This thesis aims to improve human-robot conversational groups, in which a robot is situated in an F-formation with humans. This is an important area to improve as with the increase of social robots that assist humans in hotels, shopping malls, etc., there will be more conversational interactions between humans and robots. In the mentioned scenarios, for effective interaction, the robot must know who is part of a conversational group, who is speaking, how to stand in the group, and how to interpret people's body language. To improve human-robot interaction, there is a need to understand how a robot should understand the dynamics in a conversational group and be engaged in it. To achieve these goals, this thesis introduces six contributions to the field. First, three novel datasets were collected to study human-robot and human-human conversational groups. Second, to understand the differences between conversational groups in which all the participants are human and conversational groups that contain robots, I compared the human-human and human-robot conversational groups to build knowledge of how we may adapt our human-human conversational group knowledge to adopt our human-human conversational group knowledge into human-robot conversational groups. Third, I introduce two metrics as a way to compare F-formations of the same size. Currently, F-formations are typically only described by the number of people in a group. However, group size alone may be insufficient for understanding essential differences in F-formations (e.g., sharing activity, presenting, etc.). Fourth, I present REFORM, a state-of-the-art algorithm that takes a data-driven method in detecting F-formations and improves classification accuracy beyond past approaches. Fifth, I introduce a way to detect the validity of the output of F-formation detection algorithms by looking into the way people are positioned in an F-formation. Last, but not least, I present a way to detect where a robot should look based

关键词： Visualization   Feature extraction   Semantics   Task analysis   Predictive models   Computer science   Computational modeling   Image captioning   memory initialization   recurrent neural network   visual attention   reinforcement learning  

摘要： Image captioning is a challenging task of visual understanding and has drawn more attention of researchers. In general, two inputs are required at each time step by the Long Short-Term Memory (LSTM) network used in popular attention based image captioning frameworks, including image features and previous generated words. However, error will be accumulated if the previous words are not accurate and the related semantic is not efficient enough. Facing these challenges, a novel model named CaptionNet is proposed in this work as an improved LSTM specially designed for image captioning. Concretely, only attended image features are allowed to be fed into the memory of CaptionNet through input gates. In this way, the dependency on the previous predicted words can be reduced, forcing model to focus on more visual clues of images at the current time step. Moreover, a memory initialization method called image feature encoding is designed to capture richer semantics of the target image. The evaluation on the benchmark MSCOCO and Flickr30K datasets demonstrates the effectiveness of the proposed CaptionNet model, and extensive ablation studies are performed to verify each of the proposed methods. The project page can be found in https://***/3f/9c/c9778a147356/***.

关键词： Web services   Quality of service   Predictive models   Prediction algorithms   Computer science   Time factors   Quality of service   quality prediction   matrix factorization   neighborhood information   covering algorithm  

摘要： The number of Web services on the Internet has been growing rapidly. This has made it increasingly difficult for users to find the right services from a large number of functionally equivalent candidate services. Inspecting every Web service for their quality value is impractical because it is very resource consuming. Therefore, the problem of quality prediction for Web services has attracted a lot of attention in the past several years, with a focus on the application of the Matrix Factorization (MF) technique. Recently, researchers have started to employ user similarity to improve MF-based prediction methods for Web services. However, none of the existing methods has properly and systematically addressed two of the major issues: 1) retrieving appropriate neighborhood information, i.e., similar users and services;2) utilizing full neighborhood information, i.e., both users' and services' neighborhood information. In this paper, we propose CNMF, a covering-based quality prediction method for Web services via neighborhood-aware matrix factorization. The novelty of CNMF is twofold. First, it employs a covering-based clustering method to find similar users and services, which does not require the number of clusters and cluster centroids to be prespecified. Second, it utilizes neighborhood information on both users and services to improve the prediction accuracy. The results of experiments conducted on a real-world dataset containing 1,974,675 Web service invocation records demonstrate that CNMF significantly outperforms eight existing quality prediction methods, including two state-of-the-art methods that also utilize neighborhood information with MF.

关键词： Software utilities   Stars & galaxies   Telescopes   Dark energy   Astronomy   Astrophysics   Computer science   Energy   Optics   Physics  

摘要： We examine physical properties of several galaxy clusters located within the Aquarius (AQR) and Microscopium (MSC) super-cluster complexes via their luminosity functions, fitted using Schechter functions. Images of clusters were acquired via the DECam (Dark Energy Camera) imager of the 4-meter diameter telescope at CTIO (Cerro Tololo Inter-American Observatory). We observe six clusters in our sample; four from the Aquarius supercluster (AQR 2541, 2546, 2554, 2555) and two from the Microscopium supercluster (MSC 3695, 3696). Magnitude information and density of member galaxies in each cluster were extracted via the SExtractor software utility from obtained images. We correct for background by sampling and subtracting non-cluster regions of our images. We construct luminosity functions (observations of magnitude vs cluster galaxy number density) for each cluster in our sample. Schechter functions are fitted to luminosity functions via chi-squared minimization. Fitted Schechter parameters α (characteristic slope) and M* (characteristic magnitude) of samples are compared to parameters given in other publications of clusters of different environments and redshift. Derived Schechter parameters for Aquarius clusters suggest low cluster richness, possibly due to an abundance of dwarf galaxies within the region.

关键词： Computer science  

摘要： Despite decades of research on intelligence and fundamental components of cognition, we still know very little about the structure and functionality of nervous systems. Questions in cognition and intelligent behavior are addressed by scientists in the fields of behavioral biology, neuroscience, psychology, and computer science. Yet it is difficult to reverse engineer observed sophisticated intelligent behaviors in animals and even more difficult to understand their underlying mechanisms. In this dissertation, I use a recently-developed neuroevolution platform -called Markov brain networks - in which Darwinian selection is used to evolve both structure and functionality of digital brains. I use this platform to study some of the most fundamental cognitive neural circuits: 1) visual motion detection, 2) collision-avoidance based on visual motion cues, 3) sound localization, and 4) time perception. In particular, I investigate both the selective pressures and environmental conditions in the evolution of these cognitive components, as well as the circuitry and computations behind them. This dissertation lays the groundwork for an evolutionary agent-based method to study the neural circuits for cognition in silico.

关键词： Cryptography   Security   Public key   Costs   Computer science   Big Data   Standards   Multi-key   fully homomorphic encryption   learning with errors   public key setting   ciphertext extension   distributed   circular-security   peikert   shiehian   dynamic  

摘要： Multi-key Fully homomorphic encryption (MFHE) schemes allow computation on the encrypted data under different keys. However, traditional multi-key FHE schemes based on Learning with errors (LWE) have the undesirable property that is the number of keys has to be fixed in advance. A dynamic multi-key FHE scheme is the most versatile variant which the information about the participants is not required before key generation. To support further homomorphic computation on extended ciphertexts and ciphertexts encrypted under additional keys, Peikert and Shiehian (TCC '16) proposed a leveled dynamic multi-key FHE scheme. Nevertheless, it introduces the circular-security assumption for the LWE parameters to ensure its security, which provides weaker security to the scheme. The problem of how to construct a LWE-based dynamic multi-key FHE scheme is still open. To address the above problem, in this work, we present a dynamic multi-key FHE scheme based on the LWE assumption in public key setting. The ciphertext can be extended and performed homomorphic evaluation with the ciphertexts encrypted under additional keys. Compared with current constructions, our proposed method requires fewer " local" memory and the process of ciphertext extension is distributed. Our proposed method provides a new way to extend the ciphertext such that the ciphertext homomorphism computation is more efficient. Our scheme is proven to be secure under standard LWE assumptions without using the circular-security assumption.

关键词： Neurosciences   Applied mathematics   Artificial intelligence   Physiology   Computer science   Propagation   Deep learning   Back propagation   Assignment problem   Optimization   Neural networks   Algorithms  

摘要： Synaptic plasticity is the primary physiological mechanism underlying learning in the brain. It is dependent on pre- and post-synaptic neuronal activities, and can be mediated by neuromodulatory signals. However, to date, computational models of learning that are based on pre- and post-synaptic activity and/or global neuromodulatory reward signals for plasticity have not been able to learn complex tasks that animals are capable of. In the machine learning field, neural network models with many layers of computations trained using gradient descent have been highly successful in learning difficult tasks with near-human level performance. To date, it remains unclear how gradient descent could be implemented in neural circuits with many layers of synaptic connections. The overarching goal of this thesis is to develop theories for how the unique properties of neurons can be leveraged to enable gradient descent in deep circuits and allow them to learn complex tasks. The work in this thesis is divided into three projects. The first project demonstrates that networks of cortical pyramidal neurons, which have segregated apical dendrites and exhibit bursting behavior driven by dendritic plateau potentials, can in theory leverage these physiological properties to approximate gradient descent through multiple layers of synaptic connections. The second project presents a theory for how ensembles of pyramidal neurons can multiplex sensory and learning signals using bursting and short-term plasticity, in order to approximate gradient descent and learn complex visual recognition tasks that previous biologically inspired models have struggled with. The final project focuses on the fact that machine learning models implementing gradient descent assume symmetric feedforward and feedback weights, and presents a theory for how the spiking properties of neurons can enable them to align feedforward and feedback weights in a network. As a whole, this work aims to bridge the gap between pow

关键词： Norms   Expected values   Society   Computer science   Artificial intelligence  

摘要： Norms help regulate a society. Norms can naturally emerge from interactions in a society, including conflicts between existing norms and how those conflicts are resolved. We study explicit norms, those that are represented in some structured form by the agents. Previous studies on emergence have not tackled explicit norms. We propose an approach for the emergence of explicit norms by developing agents who provide and reason about explanations for norm violations in deciding what sanctions to produce and identifying alternative norms. These agents use a genetic algorithm to create and evolve a set of norms and reinforcement learning to learn the value of these norms. We find that providing and evaluating explanations leads to the emergence of norms that provide better cohesion and goal satisfaction for member agents. Our results are stable for societies with differing attitudes of generosity.