Mini-batch stochastic gradient descent and variants thereof have become standard for large-scale empirical risk minimization like the training of neural networks. These methods are usually used with a constant batch size chosen by simple empirical inspection. The batch size significantly influences the behavior of the stochastic optimization algorithm, though, since it determines the variance o...

An advanced regression scheme is proposed to analyze fine leak batch testing data of multiple MEMS packages. The scheme employs the forward-stepwise regression method to infer the information of leaky packages from a batch test data. The analysis predicts the number of leaky packages and the true leak rate of each leaky package in a progressive manner. The scheme is implemented successfully usi...

Background Glutamine metabolism represents one of the major targets in metabolic engineering and process optimization due to its importance as cellular energy, carbon and nitrogen source. Metabolic flux analysis represents a powerful method to investigate the physiology and metabolism of cells [1]. Classical metabolic flux analysis methods require steady state conditions. However, industrially ...

In this work we compare different batch construction methods for mini-batch training of recurrent neural networks. While popular implementations like TensorFlow and MXNet suggest a bucketing approach to improve the parallelization capabilities of the recurrent training process, we propose a simple ordering strategy that arranges the training sequences in a stochastic alternatingly sorted way. W...

A simultaneous flow-injection analysis (FIA) manifold that could analyse three ions from a single injection was designed, constructed, calibrated and used successfully to analyse Li(+), Na(+) and K(+). This FIA method was 10 times faster than the batch technique. The sample volume required was a fraction of about 1/110 to 1/75 that of the batch technique. The outputs were quite reproducible and...

In this paper, we consider an integrated scheduling problem of production and distribution for manufacturers. In the production part, the batch-processing machines have fixed capacity and the jobs have arbitrary sizes and processing times. Jobs in a batch can be processed together, provided that the total size of the jobs in the batch does not exceed the machine capacity. The processing time of...

The batch-growth of linear-shaped single CNFs onto commercially available Si cantilevers (3 – 9 chips / batch) using the ion-irradiation method was challenged, and the growth parameters were optimized in terms of the sample temperature (up to 70C) and the ion-irradiation time. In every growth condition tested here, the batch-fabrication of CNF probes was achieved. The length and the growth dire...

Naproxen is a derivative of propionic acid which inhibits the cyclooxygenase enzyme and is widely used to treat patients with rheumatoid arhritis. It has antiinflammatory, analgesic and antipyretic activity. In the study, HPLC/DAD system was used for the quantitative determination of naproxen in human plasma by means of method which was developed and validated originally in Novagenix Bioanalyti...

The statistical monitoring of batch manufacturing processes is considered. It is known that conventional monitoring approaches, e.g. principal component analysis (PCA), are not applicable when the normal operating conditions of the process cannot be sufficiently represented by a Gaussian distribution. To address this issue, Gaussian mixture model (GMM) has been proposed to estimate the probabil...

This study addresses the scheduling of supply chain with interrelated factories containing suppliers and manufacturers. These elements of the chain are positioned in series and thus the efficient design of the link among them would insure good performance of the whole. In this paper, jobs transportation among factories and also delivery to the customer can be performed by batch of jobs. The cap...