The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
For the re-evolution of the mobile robot behavior in unknown environments, the mapping relation was constructed between input of sensors and output of actuators based on echo state network. An algorithm of adaptive behavior learning was presented based on echo state network for evolutionary robotics. The composite architecture with responsive behavior and behavior learning was adopted. The responsive...
Iterative learning identification algorithms for time-varying neural networks training are presented, by which neural networks based identification for discrete-time varying nonlinear systems can be carried out, as the system undertaken performs tasks repeatedly over a finite time interval. This paper develops the iterative learning least squares algorithm with dead-zone for the weights updating along...
This paper presents a new classification algorithm on traffic state of expressway which integrates the ensemble learning and fuzzy system, which consists of two fuzzy classifiers and a speed-based classifier. The fuzzy rules of two fuzzy classifiers are developed based on expert knowledge and how to optimize the parameters in fuzzy classifiers is given. While the outputs of individual classifier are...
The recovery of three dimensional structure and motion from time vary images with the aid of CCD camera(s) is usually performed using a nonlinear dynamic system, often referred to as a perspective dynamic system, where the major task is formulated as the problem of state estimation and parameter estimation. A Luenberger-type observer can be used to measure the constant motion parameter system states...
How to improve efficiency of learning is always the key issue for implementation of reinforcement learning. This paper makes use of advantages of both hierarchical learning and model-based learning, so that an improved algorithm, named Bayesian-MAXQ learning, is introduced, in which several modifications are developed to solve the value update of hierarchy, while the possible performance damages brought...
A evolutionary programming is proposed in this paper to automatically design neural networks(NNS) ensembles. Based on negative correlation learning, different individual NNs in the ensemble can learn to subdivide the task and thereby solve it more efficiently and elegantly. At the same time, different individual NNs are always to find the best collaboration connection during the evolutionary process...
A new approach called dynamic programming field for modeling the robot environments is presented and it's beneficial to the path planning. The dynamic programming field, which is approximated by Neuro-Dynamic Programming, records environmental information through a neural network and can be used to compute the approximate optimal cost between any two points. Based on the dynamic programming field,...
This paper presents the design of an iterative learning controller for a class of uncertain time-varying nonlinear systems in the presence of initial state errors. Through the introduction of initial rectified attractors and a finite-time dead-zone, a neural network iterative learning controller is designed with the proposed learning mechanism for the time-varying neural network training. The complete...
This paper presents a particle swarm optimization(PSO) algorithm in training a back propagation (BP) neural network which is used for a prediction model of the results of the college students. The present algorithm is used to combine the influencing factors of the students results with the results themselves. Compared to the BP algorithm, the present algorithm achieves better error precision with...
With regards to the petrochemical processes with various operating states and dynamic performance which will affect estimation precision for the static soft sensor, a time series soft sensor model which uses the time series of process variables to estimate the dynamic performance of quality variable was proposed. Meanwhile, the integrated Adaboost learning algorithm is introduced. With the help of...
Based on the sensitivity-based approach, we discuss the reinforcement learning problem of semi-Markov decision processes (SMDPs) with average reward. First, we provide a new Bellman optimality equation. On this basis, we propose a relative value iteration (RVI) reinforcement learning algorithm. The new RVI reinforcement learning algorithm may avoid the estimation of optimal average reward in the process...
With the rapid advancement of information technology, flood of digital data collected by business, government, and scientific applications need analyzing, digesting, and understanding. Scalability has become a necessity for data mining algorithms to process large data more effectively and extract insightful information from large data. In this paper a scaling up neural network learning algorithm is...
In MAS, model-free action-value based reinforcement learning, such as Q-learning, suffers from the fact that both the state and the action space scale exponentially with the number of agents, the learning process is very slow and low efficiency, meanwhile, the convergence of multi-agent reinforcement learning is not guaranteed when ideal assumptions do not hold. To solve the question, this paper proposes...
A RBF (Back-Propogation) neural networks based feature is applied to the target recognition, which aims at only recognition of the target feature and searches the hyperplane of the local space taking the target feature as center. The classifier integrates the target feature with RBF ANNs. It evaluates importance of each sample to the target feature by using expected output of the dynamic ANNs training...
A RBF neural network considering the critic mechanism is introduced to predict the system marginal price (SMP). The system consists of three elements, which are a predictor, an evaluator and a learning machine. The predictor is used to forecast the future SMP. The estimator is used to evaluate the prediction's validity. The explorer is used to determine the predictive step length. And the learning...
To solve the control problem of complex nonlinear system, a control strategy based on Data-drive is proposed. The method builds the local model on-line based on lazy learning algorithm, obtains the optimal control by solving the quadratic optimization problem formulated in a novel generalized predictive control framework, as a result, its adapting and anti-jamming are powerful. Further more, by using...
This paper study the multi-objective optimization problem of elevator group control systems by using the Markov Decision Process model. Define the Agent to be the leaner and decision-maker of the MDP model. And then using reinforcement learning Algorithm combined with generic method defines the elements of this model. Moreover we use SARSA(λ) value iteration algorithm which was selected to iterative...
An intelligent inhabited environment applying interconnected embedded agents by network has intelligent reasoning, planning learning, and control capabilities. Thermal and light comforts are two major control objectives for the environment to deal with using data-driven control method. Practically, dynamic association level of agents should be learned from online data with three reasons: changing...
Very short-term load forecasting predicts load over one hour into the future in five minute steps and performs the moving forecast every five minutes. This is essential for area generation control and resource dispatch, and helps operators make good decisions. To quantify prediction accuracy, it is desirable to have a confidence interval for the forecasted load in real time. However, effective prediction...
PSO has been proved as an effective supervised learning system in recent years, but it's not an effective method for incremental learning problems. Aiming at the incremental learning target for classification, a hybrid algorithm of Particle Swarm Optimization (PSO) and Artificial Immune System (AIS) called Immune based PSO (IPSO) is presented in this paper. IPSO inherits the incremental learning ability...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.