Neural machine translation (NMT) is a software that uses neural network techniques to translate text from one language to another. However, one of the most famous NMT models—Google Translate—failed to give an accurate English translation of a famous Korean nursery rhyme, "Airplane" (비행기). The authors fine-tuned a pre-trained model first with a dataset from the lyrics domain, and then with a smaller dataset containing the rhythmical properties, to teach the model to translate rhythmically accurate lyrics. This stacked fine-tuning method resulted in an NMT model that could maintain the rhythmical characteristics of lyrics during translation while single fine-tuned models failed to do so.
In this study, the authors seek to improve a machine learning algorithm used for image classification: identifying male and female images. In addition to fine-tuning the classification model, they investigate how accuracy is affected by their changes (an important task when developing and updating algorithms). To determine accuracy, a set of images is used to train the model and then a separate set of images is used for validation. They found that the validation accuracy was close to the training accuracy. This study contributes to the expanding areas of machine learning and its applications to image identification.
The purpose of the study was to determine whether graph-based machine learning techniques, which have increased prevalence in the last few years, can accurately classify data into one of many clusters, while requiring less labeled training data and parameter tuning as opposed to traditional machine learning algorithms. The results determined that the accuracy of graph-based and traditional classification algorithms depends directly upon the number of features of each dataset, the number of classes in each dataset, and the amount of labeled training data used.
Here recognizing the importance of urban green space for the health of humans and other organisms, the authors investigated if mathematical modeling can be used to develop an urban greenery management plan with high eco-sustainability by calculating the composition of a plant community. They optimized and tested their model against green fields in a Beijing city park. Although the compositions predicted by their models differed somewhat from the composition of testing fields, they conclude that by using a mathematical model such as this urban green space can be finely designed to be ecologically and economically sustainable.
Caenorhabditis elegans xpa-1 and him-1 are orthologs of human XPA and human SMC1A, respectively. Mutations in the XPA are correlated with Xeroderma pigmentosum, a condition that induces hypersensitivity to ultraviolet (UV) radiation. Alternatively, SMC1A mutations may lead to Cornelia de Lange Syndrome, a multi-organ disorder that makes patients more sensitive to UVinduced DNA damage. Both C. elegans genes have been found to be involved in protection against UV radiation, but their combined effects have not been tested when they are both knocked down. The authors hypothesized that because these genes are involved in separate pathways, the simultaneous knockdown of both of these genes using RNA interference (RNAi) in C. elegans will cause them to become more sensitive to UV radiation than either of them knocked down individually. UV protection was measured via the percent survival of C. elegans post 365 nm and 5.4x10-19 joules of UV radiation. The double xpa-1/him-1 RNAi knockdown showed a significantly reduced percent survival after 15 and 30 minutes of UV radiation relative to wild-type and xpa-1 and him-1 single knockdowns. These measurements were consistent with their hypothesis and demonstrated that xpa-1 and him-1 genes play distinct roles in resistance against UV stress in C. elegans. This result raises the possibility that the xpa-1/him-1 double knockdown could be useful as an animal model for studying the human disease Xeroderma pigmentosum and Cornelia de Lange Syndrome.
Cross country is a popular sport in the U.S. Both athletes and coaches are interested in the factors that make runners successful. In this study, the authors explore the relationship between runners' physical attributes and their race performance.
Machine learning algorithms are becoming increasingly popular for data crunching across a vast area of scientific disciplines. Here, the authors compare two machine learning algorithms with respect to accuracy and user-friendliness and find that random forest algorithms outperform logistic regression when applied to the same dataset.