We study a stochastic version of the classic orienteering problem where the time to traverse an edge is a continuous random variable. For a given temporal deadline B, our solution produces a policy, i.e., a function t...
详细信息
ISBN:
(纸本)9781665417143
We study a stochastic version of the classic orienteering problem where the time to traverse an edge is a continuous random variable. For a given temporal deadline B, our solution produces a policy, i.e., a function that, based on the current position along a solution path and the elapsed time, decides whether to continue along the path or take a shortcut to avoid missing the deadline. The solution is based on a formulation using constrained Markov decision processes to ensure that the deadline is met with a preassigned confidence level. To expedite the computation, a Monte Carlo simulation on an open loop policy is run to determine how to adaptively discretize the temporal dimension and therefore reduce the number of states and the number of optimization variables in the associated linear program. Our results show that the adaptive algorithm matches the performance of the non-adaptive one while taking significantly less time.
We propose a new adaptive algorithm for the approximation of the Landau–Lifshitz–Gilbert equation via a higher-order tangent plane scheme. We show that the adaptive approximation satisfies an energy inequality and d...
详细信息
This paper introduces the application and classification of an adaptive filtering algorithm in the image enhancement algorithm. And the filtering noise reduction impact is compared using MATLAB software for programmin...
详细信息
This paper introduces the application and classification of an adaptive filtering algorithm in the image enhancement algorithm. And the filtering noise reduction impact is compared using MATLAB software for programming, image processing, LMS algorithm, RLS algorithm, histogram equalisation algorithm, and Wiener filtering method filtering noise reduction effect. To optimize the intelligent graphic image interaction system, the proposed nonlinear adaptive algorithm of intelligent graphic image interaction system research is based on the digital filter and adaptive filtering algorithm for simulation experiment. The experimental results of several noise index data filtering algorithms show that the fuzzy coefficient k of LMS index is 0.86, RLS index is 0.91, the histogram equalization index is 0.53, and the Wiener filtering index is 0.62. LMS index of quality index Q is 0.90, RLS index is 0.95, histogram equalization index is 0.58, Wiener filtering index is 0.65. According to the above results, comparing LMS with the RLS method and according to SNR, k, and Q values in the simulation results in the process of processing, it is found that the convergence speed of the RLS algorithm is obviously better than that of the LMS algorithm, and the stability is also good. Additionally, the differential imaging data can provide a strong reference for the clinical diagnosis and qualitative differentiation of TBP and CP, and MSCT is worthy of extensive application in the clinical diagnosis of peritonitis. The processing effect of the image with high similarity to the original image is greatly improved compared with the histogram equalization and Wiener filtering methods used in the simulation.
We study a variant of stochastic bandits where the feedback model is specified by a graph. In this setting, after playing an arm, one can observe rewards of not only the played arm but also other arms that are adjacen...
详细信息
ISBN:
(纸本)9781577358664
We study a variant of stochastic bandits where the feedback model is specified by a graph. In this setting, after playing an arm, one can observe rewards of not only the played arm but also other arms that are adjacent to the played arm in the graph. Most of the existing work assumes the reward distributions are stationary over time, which, however, is often violated in common scenarios such as recommendation systems and online advertising. To address this limitation, we study stochastic bandits with graph feedback in non-stationary environments and propose algorithms with graph-dependent dynamic regret bounds. When the number of reward distribution changes L is known in advance, one of our algorithms achieves an (O) over tilde(root alpha LT) dynamic regret bound. We also develop an adaptive algorithm that can adapt to unknown L and attain an (O) over tilde(root theta LT) dynamic regret. Here, alpha and theta are some graph-dependent quantities and T is the time horizon.
Despite having several modern transformer protection schemes, the protection engineer prefers the universally adopted unit-type differential protection scheme for the current transformer (CT) because of its functional...
详细信息
In this paper, a degeneracy avoidance method for a point and line based visual SLAM algorithm is proposed. Visual SLAM predominantly uses point features. However, point features lack robustness in low texture and illu...
详细信息
ISBN:
(纸本)9781728190778
In this paper, a degeneracy avoidance method for a point and line based visual SLAM algorithm is proposed. Visual SLAM predominantly uses point features. However, point features lack robustness in low texture and illuminance variant environments. Therefore, line features are used to compensate the weaknesses of point features. In addition, point features are poor in representing discernable features for the naked eye, meaning mapped point features cannot be recognized. To overcome the limitations above, line features were actively employed in previous studies. However, since degeneracy arises in the process of using line features, this paper attempts to solve this problem. First, a simple method to identify degenerate lines is presented. In addition, a novel structural constraint is proposed to avoid the degeneracy problem. At last, a point and line based monocular SLAM system using a robust optical-flow based lien tracking method is implemented. The results are verified using experiments with the EuRoC dataset and compared with other state-of-the-art algorithms. It is proven that our method yields more accurate localization as well as mapping results.
This paper presents a fully adaptive Minimum Mean Square Error (MMSE) iterative equalization with a joint phase estimation, using various adaptive step-sizes scheme. To meet the requirement of fast convergence and low...
详细信息
ISBN:
(纸本)9780692935590
This paper presents a fully adaptive Minimum Mean Square Error (MMSE) iterative equalization with a joint phase estimation, using various adaptive step-sizes scheme. To meet the requirement of fast convergence and low MSE over time-varying channels, we propose an original self-optimized algorithm whose step-sizes are updated adaptively and assisted by soft-information provided by the channel decoder in an iterative manner. Simulation results show that our proposal achieves better performance over various multipath time-varying channels, compared to the conventional equalizer using a fixed step-size.
Modern applications in machine learning have seen more and more usage of nonconvex formulations in that they can often better capture the problem structure. One prominent example is the Deep Neural Networks which have...
详细信息
Modern applications in machine learning have seen more and more usage of nonconvex formulations in that they can often better capture the problem structure. One prominent example is the Deep Neural Networks which have achieved innumerable successes in various fields including computer vision and natural language processing. However, optimizing a non-convex problem presents much greater difficulties compared with convex ones. A vastly popular optimizer used for such scenarios is Stochastic Gradient Descent (SGD), but its performance depends crucially on the choice of its step sizes. Tuning of step sizes is notoriously laborious and the optimal choice can vary drastically across different problems. To save the labor of tuning, adaptive algorithms come to the rescue: An algorithm is said to be adaptive to a certain parameter (of the problem) if it does not need a priori knowledge of such parameter but performs competitively to those that know it. This dissertation presents our work on adaptive algorithms in following scenarios: 1. In the stochastic optimization setting, we only receive stochastic gradients and the level of noise in evaluating them greatly affects the convergence rate. Tuning is typically required when without prior knowledge of the noise scale in order to achieve the optimal rate. Considering this, we designed and analyzed noise-adaptive algorithms that can automatically ensure (near)-optimal rates under different noise scales without knowing it. 2. In training deep neural networks, the scales of gradient magnitudes in each coordinate can scatter across a very wide range unless normalization techniques, like BatchNorm, are employed. In such situations, algorithms not addressing this problem of gradient scales can behave very poorly. To mitigate this, we formally established the advantage of scale-free algorithms that adapt to the gradient scales and presented its real benefits in empirical experiments. 3. Traditional analyses in non-convex optimization
This paper focuses on the quasi-optimality of an adaptive nonconforming FEM for a distributed optimal control problem governed by the Stokes equations. The nonconforming lowest order Crouzeix-Raviart element and piece...
详细信息
To make the acoustic signal that was processed by the noise canceller sound more natural for users [1]–[3], the delay in anti-noise generation should be reduced. For a single buffer, processing delay occurrs because ...
详细信息
To make the acoustic signal that was processed by the noise canceller sound more natural for users [1]–[3], the delay in anti-noise generation should be reduced. For a single buffer, processing delay occurrs because it is impossible to write input signals while the processor is processing the data. when interfering with anti-noise and output signal, this processing delay creates additional buffering overhead to match the phase. The processing delay can be minimized using an Even-/Odd-buffer Structure to alternately read and write operations. In addition, the differences between the two methods of noise cancellation (FFT-based noise cancellation and adaptive algorithm) are compared in terms of output signal quality, processing time, and power consumption. As a result, using an Even-/Odd-buffer, reduced the processing delay of a single buffer. The FFT-based noise canceling method experienced fewer errors than the adaptive noise canceling method.
暂无评论