(1963). Cybernetics, or Control and Communication in the Animal and the Machine. Technometrics: Vol. 5, No. 1, pp. 128-130.

Cybernetics, or Control and Communication in the Animal and the Machine

As a robust nonlinear similarity measure in kernel space, correntropy has received increasing attention in domains of machine learning and signal processing. In particular, the maximum correntropy criterion (MCC) has recently been successfully applied in robust regression and filtering. The default kernel function in correntropy is the Gaussian kernel, which is, of course, not always the best choice. In this paper, we propose a generalized correntropy that adopts the generalized Gaussian density (GGD) function as the kernel, and present some important properties. We further propose the generalized maximum correntropy criterion (GMCC) and apply it to adaptive filtering. An adaptive algorithm, called the GMCC algorithm, is derived, and the stability problem and steady-state performance are studied. We show that the proposed algorithm is very stable and can achieve zero probability of divergence (POD). Simulation results confirm the theoretical expectations and demonstrate the desirable performance of the new algorithm.

Generalized Correntropy for Robust Adaptive Filtering

Maximum correntropy Kalman filter

Effective connectivity: Influence, causality and biophysical modeling

The maximum correntropy criterion (MCC) has received increasing attention in signal processing and machine learning due to its robustness against outliers (or impulsive noises). Some gradient based adaptive filtering algorithms under MCC have been developed and available for practical use. The fixed-point algorithms under MCC are, however, seldom studied. In particular, too little attention has been paid to the convergence issue of the fixed-point MCC algorithms. In this letter, we will study this problem and give a sufficient condition to guarantee the convergence of a fixed-point MCC algorithm.

Convergence of a Fixed-Point Algorithm under Maximum Correntropy Criterion

A two-dimensional, position-sensitive sensor-based system for positioning an object having six degrees of freedom in space, used for positioning of a silicon table and mask table of a lithography machine. The system comprises mainly a semiconductor laser (1), an optical fiber collimator (2), optical fibers (3, 7, 10, and 13), an optical fiber splitter (4), filter plates (6, 9, and 12), three PSD sensors (5, 8, and 11), and a signal processing system. Laser emitted by the semiconductor laser (1) is irradiated onto the optical fiber collimator (2), then transmitted respectively via three paths by the optical fiber splitter (4), and received by the PSD sensors (5, 8, and 11) after having background light filtered out by the filter plates (6, 9, and 12), while the positions of laser spots on the three PSD sensors (5, 8, and 11) are processed by the signal processing system to acquire the position and orientation of the object having six degrees of freedom in space. The positioning system uses contactless measurement, thus preventing the precision of platform movements from being affected by a cable, while at the same time providing characteristics of great positioning precision, expedited response, simple structure, and reduced costs.

Two-dimensional, position-sensitive sensor-based system for positioning object having six degrees of freedom in space

In this paper, we present a noisy version of the algebraic geometric approach of identifying parameters of discrete-time linear hybrid system. Two approximate ways of estimating hybrid parameters are considered: one is using MSE criteria, while the other is based on the information divergence that measures the distance between the error probability density function (PDF) of the identified model and the desired error PDF. A stochastic information divergence gradient algorithm is derived for the identification problem of non-gaussian system.

Information Theoretical Approach to Identification of Hybrid Systems

The invention discloses a silicon wafer platform with an anti-collision function. The silicon wafer platform comprises a silicon wafer platform body and a wire cable platform; the wire cable platform is installed at one side surface of the silicon wafer platform; the silicon wafer platform comprises three gas bags, four damper buffer elements and a gas source; the three gas bags are arranged in series, and are respectively fixed on other three side surfaces of the silicon wafer platform through gas bag supports; two adjacent gas bags are communicated with a gas pipeline through a damper buffer element; and the gas pipeline is fixed on the wire cable platform, and is communicated with the gas source. When two silicon wafer platforms collide, the gas bags of the two silicon wafer platform collide firstly, internal gases are extruded, and partial gases are exhausted out of the gas bags through the regulation of the damper buffer elements at two sides, thus not only is impact force buffered, but also the silicon wafer platforms can not rebound. Because the anti-collision structure has elasticity, compared with the mechanical anti-collision, the anti-collision structure is low in rigidity, is simple and compact in structure, and can be applied to a photoetching machine dual-platform exchange system.

Silicon wafer platform with anti-collision function

A method for measuring displacement of a large-range moving platform, comprising: arranging multiple beams of first measuring light parallel to one another and generated by an optical path distribution device and a position sensitive detector array in a certain manner, to ensure that at least one beam of first measuring light is detected by the position sensitive detector array when a moving platform is at any position of a moving area; a detection head array capable of determining whether a light beam is shaded being used for auxiliary measurement of a position of the moving platform; and determining a position of the moving platform that corresponds to the first measuring light measured by the position sensitive detector array, to calculate displacement of the moving platform. The method effectively enlarges a measurement range of the position sensitive detector array, and implements measurement of long range displacement of the moving platform.

Method for measuring displacement of large-range moving platform

Conventional cost functions of adaptive filtering are usually related to the errorpsilas dispersion, such as errorpsilas moments or errorpsilas entropy, but neglect the shape aspects (peaks, kurtosis, tails, etc.) of the error distribution. In this work, we propose a new notion of filtering (or estimation) in which the errorpsilas probability density function (PDF) is shaped into a desired one. As PDFs contain all the probabilistic information, the proposed method can be used to achieve the desired error variance or error entropy, and is expected to be useful in the complex signal processing and learning systems. In our approach, the information divergence between the actual errors and the desired errors is used as the cost function. By kernel density estimation, we derive the associated stochastic gradient algorithm for the finite impulse response (FIR) filter. Simulation results emphasize the effectiveness of this new algorithm in adaptive system training.

Jinchun Hu

Papers

Two-dimensional, position-sensitive sensor-based system for positioning object having six degrees of freedom in space

Information Theoretical Approach to Identification of Hybrid Systems

Silicon wafer platform with anti-collision function

Method for measuring displacement of large-range moving platform

Adaptive filtering for desired error distribution under minimum information divergence criterion