Influence of tire geometry on the horn effect

A reciprocal-configuration Boundary Element Method calculation of acoustic radiation characteristics has been implemented for a generic tire geometry. The influence of the geometric parameters on the radiation characteristics has been studied. The degree of amplification of noise sources on the tire belt is strongly affected by the overall tire width. In contrast, the tire radius predominantly influences the pattern of the varying amplification around the belt, rather than its absolute level. Radiusing the tire's 'shoulder' region is potentially beneficial in terms of lowering amplification levels, for a tire of fixed overall width. However, it is less effective than maintaining sharp shoulders and reducing the overall width. Thus, for an acoustically optimal belted tire, the overall width should be as small as possible, even if this leads to a larger diameter. The width should not be increased in order to accommodate a radiused crown region. Copyright © (2012) by the Institute of Noise Control Engineering (INCE).

The Random Forest Kernel and other kernels for big data from random partitions

We present Random Partition Kernels, a new class of kernels derived by demonstrating a natural connection between random partitions of objects and kernels between those objects. We show how the construction can be used to create kernels from methods that would not normally be viewed as random partitions, such as Random Forest. To demonstrate the potential of this method, we propose two new kernels, the Random Forest Kernel and the Fast Cluster Kernel, and show that these kernels consistently outperform standard kernels on problems involving real-world datasets. Finally, we show how the form of these kernels lend themselves to a natural approximation that is appropriate for certain big data problems, allowing $O(N)$ inference in methods such as Gaussian Processes, Support Vector Machines and Kernel PCA.