Adaptive parameter estimation

This paper presents a comparative study of five parameter estimation algorithms on four NLP tasks. Three of the five algorithms are wellknown in the computational linguistics community: Maximum Entropy (ME) estimation with L2 regularization, the Averaged Perceptron (AP), and Boosting. We also investigate ME estimation with L1 regularization using a novel optimization algorithm, and BLasso, which is a version of Boosting with Lasso (L1) regularization. We first investigate all of our estimators on two reranking tasks: a parse selection task and a language model (LM) adaptation task. ...
8p hongvang_1 16042013 21 3 Download

A systematic and unified presentation of the fundamentals of adaptive control theory in both continuous time and discrete time Today, adaptive control theory has grown to be a rigorous and mature discipline. As the advantages of adaptive systems for developing advanced applications grow apparent, adaptive control is becoming more popular in many fields of engineering and science. Using a simple, balanced, and harmonious style, this book provides a convenient introduction to the subject and improves one's understanding of adaptive control theory....
637p beobobeo 01082012 48 17 Download

inaccuracies lead to the deviation of operational space trajectory provided by the kinematic mapping. One method to deal with this issue can be found in an adaptive control. Xu and Gu proposed an adaptive control scheme for space robots in both joint space and operational space [Xu et al., 1992, Gu & Xu, 1993]. However, the adaptive control proposed in [Xu et al., 1992] requires perfect attitude control and the adaptive control in [Gu & Xu, 1993] is developed based on an underactuated system on the assumption that the acceleration of the basesatellite is measurable.
344p lulanphuong 23032012 52 24 Download

Knowing that the plant parameters can vary within their lower and upper bounds, this parametric uncertainty is formulated as an additive perturbation in the transfer function matrix. It is important to note that the controller be designed with respect to worst case uncertainty for each λij. This can be achieved by performing an optimization procedure given by (61) for 200 frequencies. Here an element by element uncertainty bound model is used for the characterization of upper bound of the uncertainty matrix. Then wij , which satisfies (62) for each λij is given in matrix form as,...
380p lulanphuong 22032012 96 34 Download

The noise is here assumed white with variance X, and will sometimes be restricted to be Gaussian. The last expression is in a polynomial form, whereas G, H are filters. Timevariability is modeled by timevarying parameters Bt. The adaptive filtering problem is to estimate these parametersb y an adaptive filter,
93p doroxon 12082010 44 7 Download

Adaptive filtering can be used to characterize unknown systems in timevariant environments. The main objective of this approach is to meet a difficult comprise: maximum convergence speed with maximum accuracy. Each application requires a certain approach which determines the filter structure, the cost function to minimize the estimation error, the adaptive algorithm, and other parameters; and each selection involves certain cost in computational terms, that in any case should consume less time than the time required by the application working in realtime....
162p japet75 25022013 28 7 Download

Multiuser CDMA receivers In this chapter we present a number of methods for multipleaccess interference (MAI) cancelation. MAI is produced by the presence of the other users in the network, which are located in the same bandwidth as our own signal. The common characteristic of all these schemes is some form of joint signal and parameter estimation for all signals present in the same bandwidth. It makes sense to implement this in a Base Station (BS) of a cellular system because all these signals are available there anyway.
35p khinhkha 30072010 49 6 Download

Modulation and demodulation 5.1 MAXIMUM LIKELIHOOD ESTIMATION We start again with the ML principle deﬁned in Section 3.1 of Chapter 3. After the signal despreading, vector of parameters θ to be estimated includes timing of the received symbols τ0 , phase of the received carrier θ0 , frequency offset of the received signal ν0 , amplitude of the signal A0 and data symbols an θ (τ0 , θ0 , ν0 , A0 , an ) After despreading, the narrowband signal can be represented as r(t) = s(t, θ ) + w(t) The likelihood becomes ˜ L(θ ) = C1 exp − C2...
24p khinhkha 30072010 54 8 Download

Code acquisition 3.1 OPTIMUM SOLUTION In this case, the theory starts with a simple problem where, for a received signal r(t) = s(t, θ ) + n(t), we have to estimate a generalized time invariant vector of parameters θ (frequency, phase, delay, data, . . .) of a signal s(t, θ ) in the presence of Gaussian noise ˆ n(t). The best that we can do is to ﬁnd an estimate θ of the parameter θ for which ˆ /r) is maximum; hence the name maximum aposterior the aposterior probability p(θ probability (MAP) estimate. In other words, the chosen estimate based on...
35p khinhkha 30072010 46 6 Download

CDMA network In this chapter, we initiate discussion on CDMA network capacity. The issue will be revisited again later in Chapter 13 to include additional parameters in a more comprehensive way. 8.1 CDMA NETWORK CAPACITY For initial estimation of CDMA network capacity, we start with a simple example of single cell network with n users and signal parameters deﬁned as in the list above. If αi is the power ratio of user i and the reference user with index 0, and Ni is the interference power density produced by user i deﬁned as αi = Pi /P0 , i = 1, . ....
0p khinhkha 30072010 41 6 Download

The central problemin estimation is to recover, to good accuracy, a set of unobservable parameters fromcorrupted data. Several optimization criteria have been used for estimation purposes over the years, but the most important,
40p nguyen4 17112009 51 5 Download

IMPULSIVE NOISE 12.1 12.2 12.3 12.4 12.5 12.6 12.7 Impulsive Noise Statistical Models for Impulsive Noise Median Filters Impulsive Noise Removal Using Linear Prediction Models Robust Parameter Estimation Restoration of Archived Gramophone Records Summary I mpulsive noise consists of relatively short duration “on/off” noise pulses, caused by a variety of sources, such as switching noise, adverse channel environments in a communication system, dropouts or surface degradation of audio recordings, clicks from computer keyboards, etc.
23p khinhkha 30072010 49 5 Download

To begin chapter 10, basic issues in sensorless control of induction motors are outlined. We then present example flux and speed estimators and observers and describe parameter adaptation and selfconmiissioning procedures. Finally, commercial ASDs with induction motors are reviewed.
25p gaugau1905 09122015 5 0 Download