The performance of machine learning methods is heavily dependent on the choice of data representation. In real world applications such as scene recognition problems, the widely used low-level input features can fail to explain the high-level semantic label concepts. In this work, we address this problem by proposing a novel patchbased latent variable model to integrate latent contextual represe...

For model-based 3D human pose estimation, even simple models of the human body lead to high-dimensional state spaces. Where the class of activity is known a priori, lowdimensional activity models learned from training data make possible a thorough and efficient search for the best pose. Conversely, searching for solutions in the full state space places no restriction on the class of motion to b...

Clarida, Galí and Gertler (CGG 2000), Orphanides and Williams (2005), Kim and Nelson (2006), and others have found time variation in the Fed’s “Taylor Rule” interest rate policy response function. CGG arbitrarily break their period into two fixedcoefficient 20-year subperiods, however, rather than letting the data tell them when and if any shift in the coefficients occurred. They also proxy the...

When using deep, multi-layered architectures to build generative models of data, it is difficult to train all layers at once. We propose a layer-wise training procedure admitting a performance guarantee compared to the global optimum. It is based on an optimistic proxy of future performance, the best latent marginal. We interpret autoencoders in this setting as generative models, by showing tha...

Patient satisfaction determinants in emergency departments (EDs) have for decades been heavily investigated. Despite great focus, a lack of consensus about which parameters are deemed most important remains. This study proposes an integrated framework for ED patient satisfaction, testing four key hypotheses concerning effects between the latent constructs wait time, information delivery, safety...

blank lineup (target absent lineup) A lineup in which all the members are known to be innocent . cognitive interview An interview technique designed to enhance memory and communication of events through the use of guided retrieval techniques . effective size The number of plausible members in a lineup . encoding specificity principle If cues that were present at encoding are also present at ret...

Let K be a quadratic extension of Q, B a quaternion algebra over Q and A = B ⊗Q K. Let O be a maximal order in A extending an order in B. The projective norm one group PO1 is shown to be isomorphic to the spinorial kernel group O′(L), for an explicitly determined quadratic Z-lattice L of rank four, in several general situations. In other cases, only the local structures of O and L are given at ...