检索结果-内蒙古大学图书馆

Computational Approach for Respiratory Pressure Parameters in Neonatal Ventilation

学校读者我要写书评

暂无评论

Computational Approach for Respiratory Pressure Parameters i...

Annual Siberian Russian Workshop on Electron Devices and Materials (EDM)

作者： Daria Lipchak Aleksey Chupov Anton Dolganov Aleksei Zhdanov Medical Devices R&D Bureau Joint Stock Company “Production Association “Urals Optical & Mechanical Plant” named after Mr. E. S. Yalamov” Yekaterinburg Russia Engineering School of Information Technologies Telecommunications and Control Systems Ural Federal University named after the first President of Russia B.N. Yeltsin Yekaterinburg Russia Machine Learning and Data Analytics Lab University of Erlangen-Nuremberg Erlangen Germany

In the context of mechanical ventilation, the accurate calculation of secondary pressure parameters is essential for clinicians to effectively assess a patient's clinical condition. In the case of neonatal resuscitation with the T-piece, both the inspiratory and expiratory triggers are manually operated. The device provides a constant flow at the initiation of inspiration and reduces the flow to limit output pressure according to the set peak inspiratory pressure. However, without an algorithmic classification of the inspiration and expiration pressure, data analysis may only be performed postfactum. To address the need for the separate trend analysis of PIP and PEEP in non-invasive mechanical ventilation with a T-piece, there is a requirement to develop and implement an accurate algorithm for respiratory pressure parameters calculation. In this study, we review the current state-of-the-art methods for the calculation of respiratory pressure parameters for non-invasive mechanical ventilation with a T-piece. We then present the development and implementation of an algorithm for respiratory pressure parameters calculation and discuss the results of the analysis. This algorithm is designed to provide accurate and reliable measurements of PIP and PEEP in neonatal resuscitation with the T-piece, allowing clinicians to more effectively monitor and assess a patient's clinical condition during mechanical ventilation. The results demonstrate the effectiveness of the algorithm in accurately calculating PIP and PEEP values, providing valuable insights into the patient's respiratory mechanics and helping to inform clinical decision-making.

关键词：

HIDRA: Head initialization across dynamic targets for robust architectures

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Drumond, Rafael Rego Brinkmeyer, Lukas Schmidt-Thieme, Lars Grabocka, Josif University of Hildesheim Information Systems and Machine Learning Lab

The performance of gradient-based optimization strategies depends heavily on the initial weights of the parametric model. Recent works show that there exist weight initializations from which optimization procedures can find the task-specific parameters faster than from uniformly random initializations, and that such a weight initialization can be learned by optimizing a specific model architecture across similar tasks via maml (Model-Agnostic Meta-learning). Current methods are limited to populations of classification tasks that share the same number of classes due to the static model architectures used during meta-learning. In this paper, we present hidra, a meta-learning approach that enables training and evaluating across tasks with any number of target variables. We show that Model-Agnostic Meta-learning trains a distribution for all the neurons in the output layer and a specific weight initialization for the ones in the hidden layers. Hydra explores this by learning one master neuron which is used to initialize any number of output neurons for a new task. Extensive experiments on the Miniimagenet and Omniglot data sets demonstrate that hidra improves over standard approaches while generalizing to tasks with any number of target variables. Moreover, our approach is shown to robustify low-capacity models in learning across complex tasks with a high number of classes for which regular MAML fails to learn any feasible *** Codes 68 Copyright © 2019, The Authors. All rights reserved.

关键词： Neurons

In hindsight: A smooth reward for steady exploration

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Jomaa, Hadi Samer Grabocka, Josif Schmidt-Thieme, Lars Information Systems and Machine Learning Lab University of Hildesheim Germany

In classical Q-learning, the objective is to maximize the sum of discounted rewards through iteratively using the Bellman equation as an update, in an attempt to estimate the action value function of the optimal policy. Conventionally, the loss function is defined as the temporal difference between the action value and the expected (discounted) reward, however it focuses solely on the future, leading to overestimation errors. We extend the wellestablished Q-learning techniques by introducing the hindsight factor, an additional loss term that takes into account how the model progresses, by integrating the historic temporal difference as part of the reward. The effect of this modification is examined in a deterministic continuous-state space function estimation problem, where the overestimation phenomenon is significantly reduced and results in improved stability. The underlying effect of the hindsight factor is modeled as an adaptive learning rate, which unlike existing adaptive optimizers, takes into account the previously estimated action value. The proposed method outperfoms variations of Q-learning, with an overall higher average reward and lower action values, which supports the deterministic evaluation, and proves that the hindsight factor contributes to lower overestimation errors. The mean average score of 100 episodes obtained after training for 10 million frames shows that the hindsight factor outperforms deep Q-networks, double deep Q-networks and dueling networks for a variety of ATARI games. Copyright © 2019, The Authors. All rights reserved.

关键词： Reinforcement learning

Multi-step Forecasting via Multi-task learning

学校读者我要写书评

暂无评论

Multi-step Forecasting via Multi-task Learning

IEEE International Conference on Big Data

作者： Shayan Jawed Ahmed Rashed Lars Schmidt-Thieme Information Systems and Machine Learning Lab University of Hildesheim Hildesheim Germany

ISBN: (数字)9781728108582

ISBN: (纸本)9781728108599

Multi-task learning is an established approach for improving the generalization of a model. We explore multi-task learning in the context of time series forecasting. Specifically, we look into a multivariate setting where main and auxiliary series are to be forecasted for multi-step ahead. This results in an interesting multi-task learning problem formulation where the learning tasks come from future horizon of main and auxiliary series both. Our proposed method relies firstly on enumerating multiple Convolutional network architectures to balance the number of shared and non-shared layers between different time series tasks. Also, as multi-step strategies minimize forecast errors over the complete horizon, loss functions would be at different scales based on model uncertainty for near versus distant future. For this reason we propose a factorization of the weight vector for the learning tasks with respect to their categorization of belonging to main or auxiliary series and index in future. An optimal number of shared and non-shared layers together with a novel weighted loss, results in superior performance over 2 real-world datasets compared with several baselines.

关键词： Task analysis Time series analysis Predictive models Forecasting Neural networks machine learning Data models

Multi-label network classification viaweighted personalized factorizations

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Rashed, Ahmed Grabocka, Josif Schmidt-Thieme, Lars Information Systems and Machine Learning Lab University of Hildesheim Hildesheim Germany

Multi-label network classification is a well-known task that is being used in a wide variety of web-based and non-web-based domains. It can be formalized as a multi-relational learning task for predicting nodes labels based on their relations within the network. In sparse networks, this prediction task can be very challenging when only implicit feedback information is available such as in predicting user interests in social networks. Current approaches rely on learning per-node latent representations by utilizing the network structure, however, implicit feedback relations are naturally sparse and contain only positive observed feedbacks which mean that these approaches will treat all observed relations as equally important. This is not necessarily the case in real-world scenarios as implicit relations might have semantic weights which reflect the strength of those relations. If those weights can be approximated, the models can be trained to differentiate between strong and weak relations. In this paper, we propose a weighted personalized two-stage multi-relational matrix factorization model with Bayesian personalized ranking loss for network classification that utilizes basic transitive node similarity function for weighting implicit feedback relations. Experiments show that the proposed model significantly outperforms the state-of-art models on three different real-world web-based datasets and a biology-based dataset. Copyright © 2019, The Authors. All rights reserved.

关键词： Websites

A smart filtering-based adaptive optimized link state routing protocol in flying ad hoc networks for traffic monitoring

学校读者我要写书评

暂无评论

Journal of King Saud University - Computer and information Sciences 2024年第4期36卷

作者： Hosseinzadeh, Mehdi Ali, Saqib Rahmani, Amir Masoud Lansky, Jan Nulicek, Vladimir Yousefpoor, Mohammad Sadegh Yousefpoor, Efat Darwesh, Aso Lee, Sang-Woong Institute of Research and Development Duy Tan University Da Nang Viet Nam School of Medicine and Pharmacy Duy Tan University Da Nang Viet Nam Department of Information Systems College of Economics and Political Science Sultan Qaboos University Al Khoudh Muscat Oman Future Technology Research Center National Yunlin University of Science and Technology Yunlin Taiwan Department of Computer Science and Mathematics Faculty of Economic Studies University of Finance and Administration Prague Czech Republic Center of Research and Strategic Studies Lebanese French University Kurdistan Region Iraq Department of Information Technology University of Human Development Sulaymaniyah Kurdistan region Iraq Pattern Recognition and Machine Learning Lab Gachon University 1342 Seongnamdaero Sujeonggu Seongnam 13120 South Korea

Nowadays, the use of drones as a fundamental element of smart cities has attracted the attention of many researchers to monitor and control the traffic of vehicles. Because of the high flexibility of multi-drone systems, like flying ad hoc networks (FANETs), they provide various services and improve modern life in smart cities. However, due to the unique features of FANET, especially the high speed of drones and rapid changes in network topology, communication reliability is a serious challenge in this network. Hence, traditional routing protocols, such as optimized link state routing (OLSR) scheme, cannot work well in these networks. In this paper, a smart filtering-based adaptive optimized link state routing (SFA-OLSR) scheme is proposed in FANETs. To increase adaptability to the FANET environment, SFA-OLSR provides a new solution to adjust the hello broadcast period so that each flying node specifies its broadcast period based on a new scale called cosine similarity between real and predicted positions. Furthermore, in SFA-OLSR, each flying node develops a filtering algorithm based on two parameters, namely link lifetime and remaining energy. The purpose of this algorithm is to reduce the size of the single-hop neighboring set of each flying node and minimize the search space when finding multi-point relays (MPRs). This increases the convergence speed of the algorithm. Then, SFA-OLSR exploits the sparrow search algorithm (SSA) to single out the best MPRs. This algorithm introduces a multi-objective function by focusing on three components, including energy, link lifespan, and neighbor degree. Lastly, the simulation process of SFA-OLSR is performed by the NS3 simulator. This process evaluates the performance of the proposed method and three schemes, namely Gangopadhyay et al., P-OLSR, and OLSR-ETX. These evaluations show that SFA-OLSR has a good performance in terms of three scales, namely packet delivery ratio, delay, and throughput, but its overhead is more than

关键词： Artificial intelligence (AI) Flying ad hoc networks (FANETs) Meta-heuristic algorithms Reliability Routing

Data-Driven Vehicle Trajectory Forecasting

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Jawed, Shayan Boumaiza, Eya Grabocka, Josif Schmidt-Thieme, Lars Information Systems and Machine Learning Lab University of Hildesheim Samelsonplatz 22 Hildesheim31141

An active area of research is to increase the safety of self-driving vehicles. Although safety cannot be guarenteed completely, the capability of a vehicle to predict the future trajectories of its surrounding vehicles could help ensure this notion of safety to a greater deal. We cast the trajectory forecast problem in a multi-Time step forecasting problem and develop a Convolutional neural network based approach to learn from trajectory sequences generated from completely raw dataset in real time. Results show improvement over baselines. Copyright © 2019, The Authors. All rights reserved.

关键词： Neural networks

Dataset2Vec: learning dataset meta-features

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Jomaa, Hadi S. Grabocka, Josif Schmidt-Thieme, Lars Information Systems and Machine Learning Lab University of Hildesheim Samelsonplatz 1 Hildesheim31141 Germany

machine learning tasks such as optimizing the hyper-parameters of a model for a new dataset or few-shot learning can be vastly accelerated if they are not done from scratch for every new dataset, but carry over findings from previous runs. Meta-learning makes use of features of a whole dataset such as its number of instances, its number of predictors, the means of the predictors etc., so called meta-features, dataset summary statistics or simply dataset characteristics, which so far have been hand-crafted, often specifically for the task at hand. More recently, unsupervised dataset encoding models based on variational auto-encoders have been successful in learning such characteristics for the special case when all datasets follow the same schema, but not beyond. In this paper we design a novel model, Dataset2Vec, that is able to characterize datasets with a latent feature vector based on batches and thus is able to generalize beyond datasets having the same schema to arbitrary (tabular) datasets. To do so, we employ auxiliary learning tasks on batches of datasets, esp. to distinguish batches from different datasets. We show empirically that the meta-features collected from batches of similar datasets are concentrated within a small area in the latent space, hence preserving similarity. We also show that using the dataset characteristics learned by Dataset2Vec in a state-of-the-art hyper-parameter optimization model outperforms the hand-crafted meta-features that have been used in the hyper-parameter optimization literature so far. As a result, we advance the current state-of-the-art results for hyper-parameter optimization. Copyright © 2019, The Authors. All rights reserved.

关键词： machine learning

Chameleon: learning model initializations across tasks with different schemas

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Brinkmeyer, Lukas Drumond, Rafael Rego Scholz, Randolf Grabocka, Josif Schmidt-Thieme, Lars Information Systems Machine-Learning Lab University of Hildesheim Samelsonplatz 1 Hildesheim31141 Germany

Parametric models, and particularly neural networks, require weight initialization as a starting point for gradient-based optimization. In most current practices, this is accomplished by using some form of random initialization. Instead, recent work shows that a specific initial parameter set can be learned from a population of tasks, i.e., dataset and target variable for supervised learning tasks. Using this initial parameter set leads to faster convergence for new tasks (model-agnostic meta-learning). Currently, methods for learning model initializations are limited to a population of tasks sharing the same schema, i.e., the same number, order, type and semantics of predictor and target variables. In this paper, we address the problem of meta-learning parameter initialization across tasks with different schemas, i.e., if the number of predictors varies across tasks, while they still share some variables. We propose Chameleon, a model that learns to align different predictor schemas to a common representation. We use permutations and masks of the predictors of the training tasks at hand. In experiments on real-life data sets, we show that Chameleon successfully can learn parameter initializations across tasks with different schemas providing a 26% lift on accuracy on average over random initialization and of 5% over a state-of-the-art method for fixed-schema learning model initializations. To the best of our knowledge, our paper is the first work on the problem of learning model initialization across tasks with different *** Codes 68 Copyright © 2019, The Authors. All rights reserved.

关键词： Semantics