Distributed Primal-Dual Optimization for Non-uniformly Distributed Data

Distributed primal-dual optimization has received many focuses in the past few years. In this framework, training samples are stored in multiple machines. At each round, all the machines conduct a sequence of updates based on their local data, and then the local updates are synchronized and merged to obtain the update to the global model. All the previous approaches merge the local updates by averaging all of them with a uniform weight. However, in many real world applications data are not uniformly distributed on each machine, so the uniform weight is inadequate to capture the heterogeneity of local updates. To resolve this issue, we propose a better way to merge local updates in the primal-dual optimization framework. Instead of using a single weight for all the local updates, we develop a computational efficient algorithm to automatically choose the optimal weights for each machine. Furthermore, we propose an efficient way to estimate the duality gap of the merged update by exploiting the structure of the objective function, and this leads to an efficient line search algorithm based on the reduction of duality gap. Combining these two ideas, our algorithm is much faster and more scalable than existing methods on real world problems.

Rekomendasi Objek Wisata Provinsi Jawa Barat Dengan Algoritma Branch and Bound

Tourists usually visit several places at once in time all the attractions of a visit in different location. One of the problems faced by travelers to reach the goal is the nearest path searching to a number of tourist objects. One of the algorithms used to solve the problem of line search algorithm is the nearest branch and bound. This algorithm describe the node as a point of tourist objects and lines as paths. Nodes and lines are given the values and weights for the calculation, so that from a couple of combinations and possibilities, the smallest obtained value is the closest path value can be reached. The recommendations of this line makes it easy for travelers to obtain tourism information in West Java for visiting different objects. Keywords: branch and bound algorithms, tourism object, search path, branching & bounding, graf Abstrak Wisatawan umumnya mengunjungi beberapa tempat sekaligus dalam waktu sekali kunjungan ke objek wisata yang letak lokasinya berbeda-beda. Salah satu masalah yang dihadapi wisatawan untuk mencapai tujuan adalah pencarian jalur terdekat ke sejumlah kunjungan objek wisata. Algoritma yang dapat digunakan untuk memecahkan masalah pencarian jalur terdekat adalah algoritma branch and bound. Algoritma tersebut menggambarkan simpul sebagai titik objek wisata dan garis sebagai jalur. Simpul dan garis diberi nilai dan bobot untuk perhitungan, sehingga dari beberapa kombinasi dan kemungkinan, didapatkan nilai terkecil yang merupakan nilai jalur terdekat yang dapat ditempuh. Rekomendasi jalur ini memudahkan wisatawan memperoleh informasi objek wisata di Jawa Barat untuk mengunjungi beberapa objek wisata yang berbeda-beda. Kata kunci: algoritma branch and bound, obyek wisata, pencarian jalur, pencabangan dan pembatasan, graf

Implementation issues of Yld2000-2d model under larger biaxial yield stress

In the field of sheet forming simulation, yield models serve as one of the most crucial factors for accurate computational results, and plane stress yield models have the capacity for both high efficiency and high accuracy. During recent years, applications of the Yld2000-2d model to sheet forming simulation of steel and aluminum have become increasingly popular due to its outstanding ability in describing these materials’ yield phenomena. For the computational implementation of this model, the Newton–Raphson iteration can correctly obtain the solutions of return mapping equations in most cases. However, it has been found in this work that the traditional iteration process may fall into a convergence problem when the yield stress is prominent (σb/σ0 > 1.2). To solve the new finding problem, a line search algorithm is added to the Newton–Raphson iteration process. Biaxial tension simulation results show that the line search algorithm could converge successfully even when σb/σ0 = 1.4. The simulation of the Erichsen test shows the applicability of the established Yld2000-2d model combined with a line search algorithm in the Newton–Raphson iteration process.

SSEalign: accurate function prediction of bacterial unannotated protein, based on effective training dataset

The functions of numerous bacterial proteins remain unknown because of the variety of their sequences. The performances of existing prediction methods are highly weak toward these proteins, leading to the annotation of “hypothetical protein” deposited in NCBI database. Elucidating the functions of these unannotated proteins is an urgent task in computational biology. We report a method about secondary structure element alignment called SSEalign based on an effective training dataset extracting from 20 well-studied bacterial genomes. The experimentally validated same genes in different species were selected as training positives, while different genes in different species were selected as training negatives. Moreover, SSEalign used a set of well-defined basic alignment elements with the backtracking line search algorithm to derive the best parameters for accurate prediction. Experimental results showed that SSEalign achieved 91.2% test accuracy, better than existing prediction methods. SSEalign was subsequently applied to identify the functions of those unannotated proteins in the latest published minimal bacteria genome JCVI-syn3.0. Results indicated that At least 99 proteins out of 149 unannotated proteins in the JCVI-syn3.0 genome could be annotated by SSEalign. In conclusion, our method is effective for the identification of protein homology and the annotation of uncharacterized proteins in the genome.