A Robust Solution to Variational Importance Sampling of Minimum Variance

Importance sampling is a Monte Carlo method where samples are obtained from an alternative proposal distribution. This can be used to focus the sampling process in the relevant parts of space, thus reducing the variance. Selecting the proposal that leads to the minimum variance can be formulated as an optimization problem and solved, for instance, by the use of a variational approach. Variational inference selects, from a given family, the distribution which minimizes the divergence to the distribution of interest. The Rényi projection of order 2 leads to the importance sampling estimator of minimum variance, but its computation is very costly. In this study with discrete distributions that factorize over probabilistic graphical models, we propose and evaluate an approximate projection method onto fully factored distributions. As a result of our evaluation it becomes apparent that a proposal distribution mixing the information projection with the approximate Rényi projection of order 2 could be interesting from a practical perspective.

Diffractive optics based automotive lighting system

Information projection using laser-based illumination systems in the automotive area is of keen interest to enhance communication between road users. Numerous work on laser-based front end projection employing refractive and reflective optics has been reported so far, while for rear end illumination efforts are more scarce and a different optical design concept due to limited volumetric size and field of view regulations is required. Here, we report on a new and versatile approach for a laser-based rear end lighting system for automotive application which enables projection of information or signals to support other road users. The design is based on thin diffractive optical elements projecting the desired patterns upon illumination. Also, for protection of the road users from the steering laser beam, a diffusive back projection screen is designed to project information while fulfilling both the field of view and safety requirements. The projection system is based on a periodic diffusive structure made of an array of biconic lenses with sizes in the millimeter range. The field of view (FOV) from the simulated lens arrays complies with the angular requirements set by the Economic Commission for Europe (ECE). As a proof of concept, the diffusive screen is fabricated using microfabrication technology and characterized. In future, the screen will be combined with thin diffractive optical elements to realize an entire integrated projection system.

Outlook Indikator Makro Global dan Sektor Pertanian 2016-2019

<p>Macroeconomic policies are important to consider in determining agricultural targets and policies. Thus, it is necessary to conduct an analysis of historical circumstances, current status, trends, and prospects of agricultural macro indicators. This study aims to forecast and to analyze the main macroeconomic indicators in the agricultural sector from 2016 to 2019. The study used secondary data and information. Projection was calculated using two alternative models, i.e. economic behavior and polynomial trend regression models. The results showed that after a slowdown in 2011-2014, Indonesia's economy rebounded in 2015-2016. GDP growth was expected 5,6 to 5,8% while inflation was 6,8 to 7,9% in 2016-2019. Agricultural GDP’s growth in 2016-2019 was estimated around 3,5-3,7%/year. Agricultural exports and imports were expected to increase to 10%/year and 12%/year for the period of 2016-2019, respectively. Agriculture will become the economic anchor through increases in food production and industrial commodities, as well as managing generating-inflation commodities’ prices.</p><p> </p><p>Abstrak</p><p>Kebijakan makroekonomi penting dipertimbangkan dalam menentukan target dan kebijakan sektor pertanian sehingga diperlukan suatu analisis mengenai keadaan historis, status terkini, kecenderungan yang terjadi, dan prospek indikator makro sektor pertanian. Kajian ini bertujuan untuk memproyeksi dan menganalisis indikator makro utama sektor pertanian tahun 2015-2019. Kajian ini menggunakan data dan informasi sekunder. Perhitungan proyeksi menggunakan dua alternatif yaitu melakukan estimasi dengan model perilaku ekonomi atau dengan model regresi tren polinomial. Hasil kajian menunjukkan bahwa setelah perlambatan pada 2011-2014, perekonomian Indonesia rebound pada 2015/2016. Pertumbuhan PDB diperkirakan pada kisaran 5,6-5,8% sementara inflasi umum berada pada kisaran 6,8-7,9% pada 2016-2019. Laju pertumbuhan PDB sektor pertanian pada 2016-2019 diperkirakan dalam kisaran 3,5-3,7%/tahun. Ekspor pertanian diperkirakan meningkat 10%/tahun pada 2016-2019. Impor pertanian akan meningkat 12%/tahun pada 2016-2019. Sektor pertanian akan menjadi jangkar perekonomian melalui peningkatan produksi pangan dan komoditas industri serta mengelola harga komoditas pemicu inflasi.</p>