Graph Transformer for drug response prediction

Previous models have shown that learning drug features from their graph representation is more efficient than learning from their strings or numeric representations. Furthermore, integrating multi-omics data of cell lines increases the performance of drug response prediction. However, these models showed drawbacks in extracting drug features from graph representation and incorporating redundancy information from multi-omics data. This paper proposes a deep learning model, GraTransDRP, to better drug representation and reduce information redundancy. First, the Graph transformer was utilized to extract the drug representation more efficiently. Next, Convolutional neural networks were used to learn the mutation, meth, and transcriptomics features. However, the dimension of transcriptomics features is up to 17737. Therefore, KernelPCA was applied to transcriptomics features to reduce the dimension and transform them into a dense presentation before putting them through the CNN model. Finally, drug and omics features were combined to predict a response value by a fully connected network. Experimental results show that our model outperforms some state-of-the-art methods, including GraphDRP, GraOmicDRP.

TF-IDF Method and Vector Space Model Regarding the Covid-19 Vaccine on Online News

Advances in information and technology have caused the use of the internet to be a concern of the general public. Online news sites are one of the technologies that have developed as a means of disseminating the latest information in the world. When viewed in terms of numbers, newsreaders are very sufficient to get the desired information. However, with this, the amount of information collected will result in an explosion of information and the possibility of information redundancy. The search system is one of the solutions which expected to help in finding the desired or relevant information by the input query. The methods commonly used in this case are TF-IDF and VSM (Vector Space Model) which are used in weighting to measure statistics from a collection of documents on the search for some information about the Covid 19 vaccine on kompas.com news then tokenizing it to separate the text, stopword removal or filtering to remove unnecessary words which usually consist of conjunctions and others. The next step is sentence stemming which aims to eliminate word inflection to its basic form. Then the TF-IDF and VSM calculations were carried out and the final result are news documents 3 (DOC 3) with a weight of 5.914226424; news documents 2 (DOC 2) with a weight of 1.767692186; news documents 5 (DOC 5) with weights 1.550165096; news document 4 (DOC 4) with a weight of 1.17141223;, and the last is news document 1 (DOC 1) with a weight of 0.5244103739.

Multi-Scale Guided Attention Network for Crowd Counting

The CNN-based crowd counting method uses image pyramid and dense connection to fuse features to solve the problems of multiscale and information loss. However, these operations lead to information redundancy and confusion between crowd and background information. In this paper, we propose a multi-scale guided attention network (MGANet) to solve the above problems. Specifically, the multilayer features of the network are fused by a top-down approach to obtain multiscale information and context information. The attention mechanism is used to guide the acquired features of each layer in space and channel so that the network pays more attention to the crowd in the image, ignores irrelevant information, and further integrates to obtain the final high-quality density map. Besides, we propose a counting loss function combining SSIM Loss, MAE Loss, and MSE Loss to achieve effective network convergence. We experiment on four major datasets and obtain good results. The effectiveness of the network modules is proved by the corresponding ablation experiments. The source code is available at https://github.com/lpfworld/MGANet.

Optimising aircraft taxi speed: Design and evaluation of new means to present information on a head-up display

The objective of this study was to design and evaluate new means of complying to time constraints by presenting aircraft target taxi speeds on a head-up display (HUD). Four different HUD presentations were iteratively developed from paper sketches into digital prototypes. Each HUD presentation reflected different levels of information presentation. A subsequent evaluation included 32 pilots, with varying flight experience, in usability tests. The participants subjectively assessed which information was most useful to comply with time constraints. The assessment was based on six themes including information, workload, situational awareness, stress, support and usability. The evaluation consisted of computer-simulated taxi-runs, self-assessments and statistical analysis. Information provided by a graphical vertical tape descriptive/predictive HUD presentation, including alpha-numerical information redundancy, was rated most useful. Differences between novice and expert pilots can be resolved by incorporating combinations of graphics and alpha-numeric presentations. The findings can be applied for further studies of combining navigational and time-keeping HUD support during taxi.

Information Redundancy Neglect versus Overconfidence: A Social Learning Experiment

We study social learning in a continuous action space experiment. Subjects, acting in sequence, state their beliefs about the value of a good after observing their predecessors’ statements and a private signal. We compare the behavior in the laboratory with the Perfect Bayesian Equilibrium prediction and the predictions of bounded rationality models of decision-making: the redundancy of information neglect model and the overconfidence model. The results of our experiment are in line with the predictions of the overconfidence model and at odds with the others’. (JEL C91, D12, D82, D83)

Information redundancy across spatial scales modulates early visual cortical processing

Visual images contain redundant information across spatial scales where low spatial frequency contrast is informative towards the location and likely content of high spatial frequency detail. Previous research suggests that the visual system makes use of those redundancies to facilitate efficient processing. In this framework, a fast, initial analysis of low-spatial frequency (LSF) information guides the slower and later processing of high spatial frequency (HSF) detail. Here, we used multivariate classification as well as time-frequency analysis of MEG responses to the viewing of intact and phase scrambled images of human faces to demonstrate that the availability of redundant LSF information, as found in broadband intact images, correlates with a reduction in HSF representational dominance in both early and higher-level visual areas as well as a reduction of gamma-band power in early visual cortex. Our results indicate that the cross spatial frequency information redundancy that can be found in all natural images might be a driving factor in the efficient integration of fine image details.

THE DATA DIAGNOSTIC METHOD OF IN THE SYSTEM OF RESIDUE CLASSES

The subject of the article is the development of a method for diagnosing data that are presented in the system of residual classes (SRC). The purpose of the article is to develop a method for fast diagnostics of data in the SRC when entering the minimum information redundancy. Tasks: to analyze and identify possible shortcomings of existing methods for diagnosing data in the SRC, to explore possible ways to eliminate the identified shortcomings, to develop a method for prompt diagnosis of data in SRC. Research methods: methods of analysis and synthesis of computer systems, number theory, coding theory in SRC. The following results were obtained. It is shown that the main disadvantage of the existing methods is the significant time of data diagnostics when it is necessary to introduce significant information redundancy into the non-positional code structure (NCS). The method considered in the article makes it possible to increase the efficiency of the diagnostic procedure when introducing minimal information redundancy into the NCS. The data diagnostics time, in comparison with the known methods, is reduced primarily due to the elimination of the procedure for converting numbers from the NCS to the positional code, as well as the elimination of the positional operation of comparing numbers. Secondly, the data diagnostics time is reduced by reducing the number of SRC bases in which errors can occur. Third, the data diagnostics time is reduced due to the presentation of the set of values of the alternative set of numbers in a tabular form and the possibility of sampling them in one machine cycle. The amount of additionally introduced information redundancy is reduced due to the effective use of the internal information redundancy that exists in the SRC. An example of using the proposed method for diagnosing data in SRC is given. Conclusions. Thus, the proposed method makes it possible to reduce the time for diagnosing data errors that are presented in the SRC, which increases the efficiency of diagnostics with the introduction of minimal information redundancy.