The article offers a look on the assessment methods for artificial intelligence (AI). According to the authors, the 2007 Association for the Advancement of Artificial Intelligence (AAAI) workshop has concluded that there are serious drawbacks in the way in which they assess the learning algorithms...
Dual Learning for Machine Translation 源码:https://github.com/yistLin/pytorch-dual-learning 本文作者提出了对偶学习可以更高效的利用单语数据。利用这种对偶学习,单语数据可以扮演和平行语料一样的角色,同时减少训练阶段对平行语料的依赖。具体来说,翻译模型的对偶学习可以描述成以下两个agent的游戏。 第一个agent...
In this paper, we evaluate five typical machine learning methods, namely the Naïve Bayes, Support Vector Machine, Artificial Neural Network, Random Forest and Gradient Tree Boosting, for formation lithology identification using data from the Daniudui gas field and the Hangjinqi gas field. The ...
In this paper various ensemble learning methods from machine learning and statistics are considered and applied to the customer choice modeling problem. Th... MV Wezel,R Potharst - 《European Journal of Operational Research》 被引量: 62发表: 2007年 Human-Level AI's Killer Application. Focuses ...
Most standard mapping methods require expert knowledge, supervision and fieldwork. In this study, we use optical data from the Rapid Eye satellite and topographic factors to analyze the potential of machine learning methods, i.e., artificial neural network (ANN), support vector machines (SVM) ...
Metrics for machine learning evaluation methods in cloud monitoring systems During the machine learning pipeline development, engineers need to validate the efficiency of the machine learning methods in order to assess the quality ... V Petrov,A Gennadinik,E Avksentieva - 《Proceedings of the Intern...
Keywords-software process; defect management process; sequence classification; machine learning I. INTRODUCTION problem: how to evaluate a software process performed in a specified scope? the existing software process evaluation methods often suffer from numerous limitations. ...
- 《Machine Learning》 被引量: 170发表: 2011年 Comparative Study on the Government Performance Evaluation Methods in Foreign Countries Three Es, Benchmarking and the Balanced Scorecard are three common methods of government performance evaluation in foreign countries. Focusing on the backg... Zhan ...
machine-learning approaches: multivariate logistic regression, least absolute shrinkage and selection operator–type regression (LASSO), random forest (RF), gradient boosting machine (GBM), and deep neural network (DNN). Previous studies consistently showed that these methods yield the best prediction ...
Given the aforementioned discrepancies in diffusional properties, we identify the sources of errors on migrating atoms, which are the atoms of interstitial or vacancy in the middle of the hopping from the current equilibrium site to the neighboring site (Methods). These atom hops are known as rare...