We investigate a dynamic decision making problem with constraints. The decision maker is free to take any action as long as the empirical frequency of the actions played does not violate pre-specified constraints. In ...
详细信息
We investigate a dynamic decision making problem with constraints. The decision maker is free to take any action as long as the empirical frequency of the actions played does not violate pre-specified constraints. In a case of violation the decision maker is penalized. We introduce the constrained no-regret learning model. In this model the set of alternative strategies, with which a dynamic decision policy is compared, is the set of stationary mixed actions that satisfy all the constraints. We show that there exists a strategy that satisfies the following properties: (i) it guarantees that after an unavoidable deterministic grace period, there are absolutely no violations;(ii) for an arbitrarily small constant epsilon > 0, it achieves a convergence rate of T-1-epsilon/2 , which improves the O(T-(1/3)) convergence rate of Mannor et al. (2009). (C) 2020 Elsevier B.V. All rights reserved.
Background and objective: In this study, we developed a computer controlled treadmill system using a recurrent fuzzy neural network heart rate controller (RFNNHRC). Treadmill speeds and inclines were controlled by cor...
详细信息
Background and objective: In this study, we developed a computer controlled treadmill system using a recurrent fuzzy neural network heart rate controller (RFNNHRC). Treadmill speeds and inclines were controlled by corresponding control servo motors. The RFNNHRC was used to generate the control signals to automatically control treadmill speed and incline to minimize the user heart rate deviations from a preset profile. Methods: The RFNNHRC combines a fuzzy reasoning capability to accommodate uncertain information and an artificial recurrent neural network learning process that corrects for treadmill system nonlinearities and uncertainties. Treadmill speeds and inclines are controlled by the RFNNHRC to achieve minimal heart rate deviation from a pre-set profile using adjustable parameters and an on-line learning algorithm that provides robust performance against parameter variations. The on-line learning algorithm of RFNNHRC was developed and implemented using a dsPIC 30F4011 DSP. Results: Application of the proposed control scheme to heart rate responses of runners resulted in smaller fluctuations than those produced by using proportional integra control, and treadmill speeds and inclines were smoother. The present experiments demonstrate improved heart rate tracking performance with the proposed control scheme. Conclusions: The RFNNHRC scheme with adjustable parameters and an on-line learning algorithm was applied to a computer controlled treadmill system with heart rate control during treadmill exercise. Novel RFNNHRC structure and controller stability analyses were introduced. The RFNNHRC were tuned using a Lyapunov function to ensure system stability. The superior heart rate control with the proposed RFNNHRC scheme was demonstrated with various pre-set heart rates. (C) 2016 Elsevier Ireland Ltd. All rights reserved.
RBF neural network off-linelearningalgorithm can only be trained by a set of samples at the same time,not by individual samples one by *** its adaptiveness is *** paper prensents an improved RBF network on-line lear...
详细信息
ISBN:
(纸本)9781424463251;9780769539911
RBF neural network off-linelearningalgorithm can only be trained by a set of samples at the same time,not by individual samples one by *** its adaptiveness is *** paper prensents an improved RBF network on-line learning algorithm based on resource allocation network-The network can be trained by individual samples *** fisrt a structure of RBF neural network with no hidden layer is created,then this network is trained using individual samples,according to current errors between output and *** numbers and locations of the hidden layer nodes are dynamically added or deleted to optimize the structure of *** strategy of increasing or decreasing nodes on-line is based on the recursive least squares *** the samples producing larger output *** weights of neurons are retained otherwise deleted to improve RBF neural network convergence speed and real time. Nonlinear curve fitting tests including natural index, multiplication, trigonometric functions and gas content forecast are carred out by VC++ simulation software to verify the validity of this *** tests show that the improved onlinelearningalgorithm for RBF network has higher forecasting accuracy,well generalization ability,fewer hidden nodes. It can be realized in embedded systems and has a good value in many application fields such as gas content forecast.
The structure of dynamic neural networks and their on-line learning algorithm are two important factors for NAICSs (nonlinear adaptive inverse control system). Recurrent neural networks are effective identifiers for n...
详细信息
ISBN:
(纸本)9787900719706
The structure of dynamic neural networks and their on-line learning algorithm are two important factors for NAICSs (nonlinear adaptive inverse control system). Recurrent neural networks are effective identifiers for nonlinear plant. However, they have complex architecture. In order to simplify the structure of dynamic neural networks, this paper proposed an improved DAFNN (dynamic activation function neural network) by changing the neurons' structure. The modified DFANN with the same simple feedforward architecture as the origin one had better dynamic ability. Gradient. vectors calculation becomes the chief bottleneck for learningalgorithms of DAFNNs due to its complex chain rule expansions. A new on-line learning algorithm was proposed to simplify the proceeding of gradient calculation for DAFNNs in this paper according to signal flow graph theory. The new algorithm had adaptive learning rates to guarantee its convergence. The algorithm was used as the learningalgorithms of identifier and controller in NAICSs. Simulation results show it is an efficient algorithm for NAICSs.
This paper presents a new algorithm for on-line artificial neural networks (ANN) training. The network topology is a standard multilayer perceptron (MLP) and the training algorithm is based on the theory of variable s...
详细信息
This paper presents a new algorithm for on-line artificial neural networks (ANN) training. The network topology is a standard multilayer perceptron (MLP) and the training algorithm is based on the theory of variable structure systems (VSS) and sliding mode control (SMC). The main feature of this novel procedure is the adaptability of the gain (learning rate), which is obtained from sliding mode surface so that system stability is guaranteed. (c) 2007 Elsevier B.V. All rights reserved.
作者：
Ratsaby, JUCL
Dept Comp Sci London WC1E 6BT England
Consider the pattern recognition problem of learning multicategory classification from a labeled sample, for instance, the problem of learning character recognition where a category corresponds to an alphanumeric lett...
详细信息
Consider the pattern recognition problem of learning multicategory classification from a labeled sample, for instance, the problem of learning character recognition where a category corresponds to an alphanumeric letter. The classical theory of pattern recognition assumes labeled examples appear according to the unknown underlying pattern-class conditional probability distributions where the pattern classes are picked randomly according to their a priori probabilities. In this paper we pose the following question: Can the learning accuracy be improved if labeled examples are independently randomly drawn according to the underlying class conditional probability distributions but the pattern classes are chosen not necessarily according to their a priori probabilities? We answer this in the affirmative by showing that there exists a tuning of the sub-sample proportions which minimizes a loss criterion. The tuning is relative to the intrinsic complexity of the Bayes-classifier. As this complexity depends on the underlying probability distributions which are assumed to be unknown, we provide an algorithm which learns the proportions in an on-line manner utilizing sample querying which asymptotically minimizes the criterion. In practice, this algorithm may be used to boost the performance of existing learning classification algorithms by apportioning better sub-sample proportions. (C) 2003 Elsevier Science (USA). All rights reserved.
A newly designed driving circuit for the traveling-wave type ultrasonic motor (USM). Which is composed of a push-pull de-dr power converter using PWM direct duty-cycle control and a current-source two-phase parallel-r...
详细信息
A newly designed driving circuit for the traveling-wave type ultrasonic motor (USM). Which is composed of a push-pull de-dr power converter using PWM direct duty-cycle control and a current-source two-phase parallel-resonant capacitor-parallel load inverter, is described in this study. Moreover, since the dynamic characteristics of the USM are difficult to obtain and the motor parameters are time varying, an on-line trained neural-network (NN) controller is proposed to control the rotor speed and position of the USM. Accurate tracking response can be obtained due to the powerful on-linelearning capability of the NN controller, Furthermore, the influence of parameter variations and external disturbances on the USM drive also can be reduced effectively. The effectiveness of the NN controlled USM drive system is demonstrated by experimental results.
暂无评论