A bayesian nonparametric joint factor model for learning shared and individual subspaces from multiple data sources

Joint analysis of multiple data sources is becoming increasingly popular in transfer learning, multi-task learning and cross-domain data mining. One promising approach to model the data jointly is through learning the shared and individual factor subspaces. However, performance of this approach depends on the subspace dimensionalities and the level of sharing needs to be specified a priori. To this end, we propose a nonparametric joint factor analysis framework for modeling multiple related data sources. Our model utilizes the hierarchical beta process as a nonparametric prior to automatically infer the number of shared and individual factors. For posterior inference, we provide a Gibbs sampling scheme using auxiliary variables. The effectiveness of the proposed framework is validated through its application on two real world problems - transfer learning in text and image retrieval.

Managing chondral lesions of the glenohumeral joint

With regard to the use of this technique in the shoulder there has been little published, and the paper by Snow on microfracture of chondral lesions of the glenohumeral joint in this edition of the IJSS aims to provide a little more data regarding outcomes. [4] PÎ ́ssler, in the German literature in 2000, described the use of microfracture in the shoulder in discussing 162 patients who were followed up following microfracture arthroplasty of a range of joints and reported overall good results, particularly for pain relief at 3-6 years.

Effects of tool positioning on joint interface microstructure and fracture strength of friction stir lap Al-to-steel welds

Friction stir lap welding (FSLW) experiments have been conducted to study the effects of tool positioning on microstructures formed in the Al-to-steel interface region and on joint strength, defined as maximum applied force over the width (F m/w s) of the test sample, of the welds. Various pin positioning and speed conditions were used in the FSLW experiments followed by microstructure examination on the interface regions and tensile-shear testing on the welds, including an examination on crack propagation in mixed stir zone. It was found that when the pin was close to the bottom steel piece, Al-to-steel reaction occurred resulting in intermetallic outbursts formed along the interface. This represents the case of incomplete metallurgical joint. When the pin was lowered to just reach the steel, a thin and continued interface intermetallic layer formed. Evidences and consideration on growth kinetics have suggested that the layer could only remain thin (≤2.5 μm) during FSLW. This layer could bear a high load during tensile-shear testing and the adjacent aluminium deformed and fractured instead. The resulting F m/w s was high. When the pin penetrated to steel, F m/w s reduced due to brittle fracture being dominant inside mixed stir zone. Evidences have shown that the amount of penetration and speed condition during FSLW do not have large effects on F m/w s.

Guided wave based methods for damage detection : experimental study on concrete joint

Since Guided wave (GW) is sensitive to small damage and can propagate a relatively longer distance with relatively less attenuation, GW-based method has been found as an effective and efficient way to detect incipient damages. In this study, a full-scale concrete joint was constructed to further verify the effectiveness of GW-based method on real civil structures. GW tests were conducted in three stages, including baseline, serviceability and damage conditions. The waves are excited by one actuator and received by several sensors, which are made up of independent piezoelectric elements. Experimental results show that the mehod is promising for damage identification in practices.

Surgical treatment of lateral clavicle fractures associated with complete coracoclavicular ligament disruption: clinico-radiological outcomes of acromioclavicular joint sparing and spanning implants

Distal clavicle fracture associated with complete coracoclavicular ligament disruption represents an unstable injury, and osteosynthesis is recommended. This study was performed (1) to retrospectively analyse the clinico-radiological outcomes of two internal fixation techniques, and (2) to identify and analyse radiographic fracture patterns of fracture that are associated with this injury. Conclusions: Internal fixation of this fracture pattern is associated with a high union rate and favorable clinical outcomes with both techniques. A combination of distal radius plate and ligament reconstruction device resulted in stable fixation and significantly lower reoperation rates, and should be used when fracture geometry permits (Types 1 and 2).

Combined joint fusion for index and middle carpometacarpal instability in elite boxers

Isolated instability of the index and middle finger carpometacarpal joints is uncommon. An unpublished injury pattern in a consecutive series of 13 elite boxers is described, with mid-term radiological and functional results of two novel surgical treatments. All the boxers (from Australia and the UK) were unable to compete owing to pain and weakness. Four boxers were managed initially by open reduction and temporary K-wire fixation. Nine boxers underwent primary carpometacarpal arthrodesis. All were able to return to their previous level of competition. One boxer who had undergone a soft tissue reconstruction competed at international level but required an arthrodesis because of recurrent symptoms. In elite boxers, simple reduction and wiring may be appropriate for an acute injury causing index or middle finger carpometacarpal joint instability, however, arthrodesis is the treatment of choice when instability and degenerative changes are present.

Dive angles for a green turtle (Chelonia mydas)

The depth and swim speed of a green turtle (Chelonia mydas) were measured during the internesting period in Cyprus. For dives to the seabed (U-dives) we used these data to determine dive angles. Typically the turtle initially descended at a steep angle ([similar]60°) but as the dive continued this angle lessened until the turtle approached the seabed at an average angle of [similar]15°. This systematic change in descent angle is consistent with the prediction that the energetic implications of dive angle are most important at the start of the dive when the turtle is fighting to overcome its positive buoyancy. On leaving the seabed, the turtle often seemed to rise passively.

Learning over sets using boosted manifold principal angles (BoMPA)

In this paper we address the problem of classifying vector sets. We motivate and introduce a novel method based on comparisons between corresponding vector subspaces. In particular, there are two main areas of novelty: (i) we extend the concept of principal angles between linear subspaces to manifolds with arbitrary nonlinearities; (ii) it is demonstrated how boosting can be used for application-optimal principal angle fusion. The strengths of the proposed method are empirically demonstrated on the task of automatic face recognition (AFR), in which it is shown to outperform state-of-the-art methods in the literature.