Extending noise robust structured support vector machines to larger vocabulary tasks

This paper describes a structured SVM framework suitable for noise-robust medium/large vocabulary speech recognition. Several theoretical and practical extensions to previous work on small vocabulary tasks are detailed. The joint feature space based on word models is extended to allow context-dependent triphone models to be used. By interpreting the structured SVM as a large margin log-linear model, illustrates that there is an implicit assumption that the prior of the discriminative parameter is a zero mean Gaussian. However, depending on the definition of likelihood feature space, a non-zero prior may be more appropriate. A general Gaussian prior is incorporated into the large margin training criterion in a form that allows the cutting plan algorithm to be directly applied. To further speed up the training process, 1-slack algorithm, caching competing hypothesis and parallelization strategies are also proposed. The performance of structured SVMs is evaluated on noise corrupted medium vocabulary speech recognition task: AURORA 4. © 2011 IEEE.

Auto-calibration of a camera system using Image Alignment

Calibration of a camera system is a necessary step in any stereo metric process. It correlates all cameras to a common coordinate system by measuring the intrinsic and extrinsic parameters of each camera. Currently, manual calibration of a camera system is the only way to achieve calibration in civil engineering operations that require stereo metric processes (photogrammetry, videogrammetry, vision based asset tracking, etc). This type of calibration however is time-consuming and labor-intensive. Furthermore, in civil engineering operations, camera systems are exposed to open, busy sites. In these conditions, the position of presumably stationary cameras can easily be changed due to external factors such as wind, vibrations or due to an unintentional push/touch from personnel on site. In such cases manual calibration must be repeated. In order to address this issue, several self-calibration algorithms have been proposed. These algorithms use Projective Geometry, Absolute Conic and Kruppa Equations and variations of these to produce processes that achieve calibration. However, most of these methods do not consider all constraints of a camera system such as camera intrinsic constraints, scene constraints, camera motion or varying camera intrinsic properties. This paper presents a novel method that takes all constraints into consideration to auto-calibrate cameras using an image alignment algorithm originally meant for vision based tracking. In this method, image frames are taken from cameras. These frames are used to calculate the fundamental matrix that gives epipolar constraints. Intrinsic and extrinsic properties of cameras are acquired from this calculation. Test results are presented in this paper with recommendations for further improvement.

Optical trapping of nanotubes with cylindrical vector beams

We use laser beams with radial and azimuthal polarization to optically trap carbon nanotubes. We measure force constants and trap parameters as a function of power showing improved axial trapping efficiency with respect to linearly polarized beams. The analysis of the thermal fluctuations highlights a significant change in the optical trapping potential when using cylindrical vector beams. This enables the use of polarization states to shape optical traps according to the particle geometry, as well as paving the way to nanoprobe-based photonic force microscopy with increased performance compared to a standard linearly polarized configuration. © 2012 Optical Society of America.

Generalized vector control for brushless doubly fed machines with nested-loop rotor

This paper presents a generalized vector control system for a generic brushless doubly fed (induction) machine (BDFM) with nested-loop type rotor. The generic BDFM consists of p1/p2 pole-pair stator windings and a nested-loop rotor with N number of loops per nest. The vector control system is derived based on the basic BDFM equation in the synchronous mode accompanied with an appropriate synchronization approach to the grid. An analysis is performed for the vector control system using the generic BDFM vector model. The analysis proves the efficacy of the proposed approach in BDFM electromagnetic torque and rotor flux control. In fact, in the proposed vector control system, the BDFM torque can be controlled very effectively promising a high-performance BDFM shaft speed control system. A closed-loop shaft speed control system is composed based on the presented vector control system whose performance is examined both in simulations and experiments. The results confirm the high performance of the proposed approach in BDFM shaft speed control as well as a very close agreement between the simulations and experiments. Tests are performed on a 180-frame prototype BDFM. © 2012 IEEE.

Structure Discovery in Nonparametric Regression through Compositional Kernel Search

Despite its importance, choosing the structural form of the kernel in nonparametric regression remains a black art. We define a space of kernel structures which are built compositionally by adding and multiplying a small number of base kernels. We present a method for searching over this space of structures which mirrors the scientific discovery process. The learned structures can often decompose functions into interpretable components and enable long-range extrapolation on time-series datasets. Our structure search method outperforms many widely used kernels and kernel combination methods on a variety of prediction tasks.

Convex relaxation of mixture regression with efficient algorithms

We develop a convex relaxation of maximum a posteriori estimation of a mixture of regression models. Although our relaxation involves a semidefinite matrix variable, we reformulate the problem to eliminate the need for general semidefinite programming. In particular, we provide two reformulations that admit fast algorithms. The first is a max-min spectral reformulation exploiting quasi-Newton descent. The second is a min-min reformulation consisting of fast alternating steps of closed-form updates. We evaluate the methods against Expectation-Maximization in a real problem of motion segmentation from video data.