Reinforcement Programming (RP) is a new approach to automatically generating algorithms, that uses reinforcement learning techniques. This paper describes the RP approach and gives results of experiments using RP to g...
详细信息
Reinforcement Programming (RP) is a new approach to automatically generating algorithms, that uses reinforcement learning techniques. This paper describes the RP approach and gives results of experiments using RP to generate a generalized, in-place, iterative sort algorithm. The RP approach improves on earlier results that that use genetic programming (GP). The resulting algorithm is a novel algorithm that is more efficient than comparable sorting routines. RP learns the sort in fewer iterations than GP and with fewer resources. Results establish interesting empirical bounds on learning the sort algorithm: A list of size 4 is sufficient to learn the generalized sort algorithm. The training set only requires one element and learning took less than 200,000 iterations. RP has also been used to generate three binary addition algorithms: a full adder, a binary incrementer, and a binary adder.
In this paper, the author used K-means and fuzzy K-means to analyze the classification of precipitation in JingDeZhen City, and the results showed that using fuzzy k-means algorithm is a more efficient data clustering...
详细信息
In this paper, the author used K-means and fuzzy K-means to analyze the classification of precipitation in JingDeZhen City, and the results showed that using fuzzy k-means algorithm is a more efficient data clustering algorithm, with better value of promotion and practical application.
The social cognitive optimization algorithm is one of the newest intelligent algorithms, and this algorithm can help the solvers to avoid tripping in local optimization when solving the nonlinear constraint problems e...
详细信息
The social cognitive optimization algorithm is one of the newest intelligent algorithms, and this algorithm can help the solvers to avoid tripping in local optimization when solving the nonlinear constraint problems effectively. The algorithm is based on the social cognitive theory and the key point of the ergodicity is the process of refreshing the knowledge points. Modified and optimized the conditions of neighborhood searching through bring in the Chaos and Kent mapping function to get more reasonable knowledge points which were distributed more uniform. Used the real nonlinear constraint problem to test the performance of the modified algorithm, through compared data, the later has advantages on the speed of convergence and the legitimacy, and the value of the target function is more closed to the theory value.
This paper proposes an improved FCM algorithm aiming at many problems in Fuzzy C Means algorithm, such as being sensitive to initial conditions, usually leading to local minimum results. The new algorithm can obtain g...
详细信息
This paper proposes an improved FCM algorithm aiming at many problems in Fuzzy C Means algorithm, such as being sensitive to initial conditions, usually leading to local minimum results. The new algorithm can obtain global optimal solutions through a new simple and efficient selecting rule of the initial cluster centers, furthermore alternating optimization in terms of a novel separable criterion. By comparative testing with custom FCM, the new algorithms not only have fewer numbers of iterations and have higher accuracy, but also more suitable for problems with not balanced classified samples. Finally, the new algorithm is applied in traffic condition recognition and the result shows that the new clustering approach is promising for the dynamic identification of road traffic state.
In this paper, an iterative method is proposed to solve the second-order Sylvester matrix equation EXF 2 +AXF+CX+BY=D with unknown matrix pair [X, Y], based on a matrix form of LSQR algorithm. By this iterative method...
详细信息
In this paper, an iterative method is proposed to solve the second-order Sylvester matrix equation EXF 2 +AXF+CX+BY=D with unknown matrix pair [X, Y], based on a matrix form of LSQR algorithm. By this iterative method, we can obtain the minimum Frobenius norm solution pair or the minimum Frobenius norm least squares solution pair over some constrained matrices, such as symmetric, generalized bisymmetric and (R, S)-symmetric matrices.
To address the unconstrained optimization problem, the Conjugate Gradient Method (CG) uses the sequence of iterations to approach the minimum point of aim function. Because of the effect of rounding errors, many merit...
详细信息
To address the unconstrained optimization problem, the Conjugate Gradient Method (CG) uses the sequence of iterations to approach the minimum point of aim function. Because of the effect of rounding errors, many merits of CG are no longer in existence in practical use. Hence the rate of convergence is not ideal and a practical problem confronting us is how to improve conjugate gradient iteration so as to accelerate the convergence. Common improvements include better descent directions and restart strategies on the precondition of conjugate gradients. From the angle of the search step length, another major factor that influences the rate of convergence, the author proposes the use of the neural network model to introduce ‘priori knowledge’ in CG so that it may predict the next search step length. Large quantities of experimental data prove that this method can effectively improve the rate of convergence.
In data mining, SVD is a popular method that has been used for compressing high dimensional data. Binary matrix factorization (BMF) is a variant of SVD. There are two methods for binary factorization compression: the ...
详细信息
In data mining, SVD is a popular method that has been used for compressing high dimensional data. Binary matrix factorization (BMF) is a variant of SVD. There are two methods for binary factorization compression: the iterative heuristic and greedy algorithms. However, both of them are not perfect in applications. The iterative heuristic does not guarantee the convergence in most cases and greedy algorithms can't fit the need of large-scale matrices factorization. In this paper a new method is used for BMF: consensus algorithms. Consensus algorithms are a brand-new approach to enumerating all the maximal bicliques for a given graph, which is proved to be an NP-complete problem and can give the solution in incremental polynomial time. For some bipartite graphs, the time complexity is polynomial. Experiments show that when the iterative heuristic does not work, consensus algorithm improves far more badly the efficiency than greedy algorithms, and ensures the stability.
A new iterative procedure for the design of linear aperiodic arrays is introduced which permits exploiting, in a combined way, both the positions and the excitation amplitudes of the array elements obtaining a pattern...
详细信息
ISBN:
(纸本)9781424451272
A new iterative procedure for the design of linear aperiodic arrays is introduced which permits exploiting, in a combined way, both the positions and the excitation amplitudes of the array elements obtaining a pattern which optimally fits, in terms of a weighted L2 norm, the pattern of a reference linear continuous aperture. Interesting radiative properties and some advantages for the realization of active arrays will be presented as well.
The paper developed a block-wise approach for ICA algorithms which can improve the computational efficiency of ICA without the degradation of performance for the separation of biomedical signals. Source signals includ...
详细信息
The paper developed a block-wise approach for ICA algorithms which can improve the computational efficiency of ICA without the degradation of performance for the separation of biomedical signals. Source signals including electrocardiogram (ECG), electromyogram (EMG) and 60-Hz sinusoid are linearly mixed for experimental tests. The mean-square errors (MSE) between the original sources and the separated signals are calculated for the evaluation of separation performance. These results demonstrated that the proposed block-wise approach can achieve the desired separation performance of signals in a more efficient way.
A robust iterative learning control algorithm is proposed based on T-S model for a kind of nonlinear time-delay systems with repetitive actions. Firstly, the proposed algorithm uses fuzzy T-S model to build the model ...
详细信息
A robust iterative learning control algorithm is proposed based on T-S model for a kind of nonlinear time-delay systems with repetitive actions. Firstly, the proposed algorithm uses fuzzy T-S model to build the model for nonlinear time-delay system, secondly, the global fuzzy system model could be described as the form of uncertain systems, finally, the robust iterative learning controller is designed through resolving Riccati equation. The sufficient and essential condition of the algorithm is deduced based on Lyapunov theories for nonlinear time-delay system.
暂无评论