Previous results have shown disappointing performance when porting a parser trained on one domain to another domain where only a small amount of data is available. We propose the use of data-defined kernels as a way to exploit statistics from a source domain while still specializing a parser to a target domain. A probabilistic model trained on the source domain (and possibly also the target dom...

This paper details a systematic method for significantly improving the noise margins of very fast threshold gates. The method is based on adding nonlinear terms determined from the Boolean form of the threshold function to be implemented. Simulation results support our theoretical claims. Finally, two methods for drastically reducing the dissipated power of such threshold gates down to <50%, an...

Mobile manipulators operating in field environments will be required to apply large forces, or manipulate large loads, and to perform such tasks on uneven terrain which may cause the system to approach, or reach, a dangerous tipover instability. To avoid tipover in an automatic system, or to provide a human operator with an indication of proximity to tipover, it is necessary to define a measure...

Margin theory provides one of the most popular explanations to the success of AdaBoost, where the central point lies in the recognition that margin is the key for characterizing the performance of AdaBoost. This theory has been very influential, e.g., it has been used to argue that AdaBoost usually does not overfit since it tends to enlarge the margin even after the training error reaches zero....

The position and intensity of the southern westerly wind belt varies seasonally as a consequence of changes in sea surface temperature. During the austral winter, the belt expands northward and the wind intensity in the core decreases. Conversely, during the summer, the belt contracts, and the intensity within the core is strengthened. Reconstructions of the westerly winds since the last glacia...

The sinking of lithosphere at subduction zones couples Earth’s exterior with its interior, spawns continental crust and powers a tectonic regime that is unique to our planet. In spite of its importance, it is unclear how subduction is initiated. Two general mechanisms are recognized: induced and spontaneous nucleation of subduction zones. Induced nucleation (INSZ) responds to continuing plate c...

This study attempts to assess the financial performance of a supply chain company. In order achieve objective, this has analysed ratios applying DuPont analysis, which been demonstrated with tables show company's trend for five years (2018-2022). analysis is based on Return Equity (ROE) and Assets (ROA). The ROE disaggregates into three components: • Net Profit Margin, Total Asset Turnover, Mul...

Recently, the need to increase the strength of reinforced concrete members has become a subject that civil engineers are interested in tackling. Of the many proposed solutions, fiber-reinforced polymer (FRP) materials have attracted attention due to their superior properties, such as high strength-to-weight ratio, high energy absorption and excellent corrosion resistance. FRP wrapping of concre...

Support vector machines (SVMs) and Boosting are possibly the two most popular learning approaches during the past two decades. It is well known that the margin is a fundamental issue of SVMs, whereas recently the margin theory for Boosting has been defended, establishing a connection between these two mainstream approaches. The recent theoretical results disclosed that the margin distribution r...

Although fuzzy support vector machine introduces the fuzzy membership degree in maximizing the margin and improves performance of classifier, it has not fully considered the position of training samples in the margin. In this paper, a double margin (rough margin) based fuzzy support vector machine (RFSVM) algorithm is presented by introducing rough set into fuzzy support vector machine. Firstly...