-
A Multiple Choices Reading Comprehension Corpus for Vietnamese Language Education
Authors:
Son T. Luu,
Khoi Trong Hoang,
Tuong Quang Pham,
Kiet Van Nguyen,
Ngan Luu-Thuy Nguyen
Abstract:
Machine reading comprehension has been an interesting and challenging task in recent years, with the purpose of extracting useful information from texts. To attain the computer ability to understand the reading text and answer relevant information, we introduce ViMMRC 2.0 - an extension of the previous ViMMRC for the task of multiple-choice reading comprehension in Vietnamese Textbooks which conta…
▽ More
Machine reading comprehension has been an interesting and challenging task in recent years, with the purpose of extracting useful information from texts. To attain the computer ability to understand the reading text and answer relevant information, we introduce ViMMRC 2.0 - an extension of the previous ViMMRC for the task of multiple-choice reading comprehension in Vietnamese Textbooks which contain the reading articles for students from Grade 1 to Grade 12. This dataset has 699 reading passages which are prose and poems, and 5,273 questions. The questions in the new dataset are not fixed with four options as in the previous version. Moreover, the difficulty of questions is increased, which challenges the models to find the correct choice. The computer must understand the whole context of the reading passage, the question, and the content of each choice to extract the right answers. Hence, we propose the multi-stage approach that combines the multi-step attention network (MAN) with the natural language inference (NLI) task to enhance the performance of the reading comprehension model. Then, we compare the proposed methodology with the baseline BERTology models on the new dataset and the ViMMRC 1.0. Our multi-stage models achieved 58.81% by Accuracy on the test set, which is 5.34% better than the highest BERTology models. From the results of the error analysis, we found the challenge of the reading comprehension models is understanding the implicit context in texts and linking them together in order to find the correct answers. Finally, we hope our new dataset will motivate further research in enhancing the language understanding ability of computers in the Vietnamese language.
△ Less
Submitted 31 March, 2023;
originally announced March 2023.
-
Optimal Design of Electric Machine with Efficient Handling of Constraints and Surrogate Assistance
Authors:
Bhuvan Khoshoo,
Julian Blank,
Thang Q. Pham,
Kalyanmoy Deb,
Shanelle N. Foster
Abstract:
Electric machine design optimization is a computationally expensive multi-objective optimization problem. While the objectives require time-consuming finite element analysis, optimization constraints can often be based on mathematical expressions, such as geometric constraints. This article investigates this optimization problem of mixed computationally expensive nature by proposing an optimizatio…
▽ More
Electric machine design optimization is a computationally expensive multi-objective optimization problem. While the objectives require time-consuming finite element analysis, optimization constraints can often be based on mathematical expressions, such as geometric constraints. This article investigates this optimization problem of mixed computationally expensive nature by proposing an optimization method incorporated into a popularly-used evolutionary multi-objective optimization algorithm - NSGA-II. The proposed method exploits the inexpensiveness of geometric constraints to generate feasible designs by using a custom repair operator. The proposed method also addresses the time-consuming objective functions by incorporating surrogate models for predicting machine performance. The article successfully establishes the superiority of the proposed method over the conventional optimization approach. This study clearly demonstrates how a complex engineering design can be optimized for multiple objectives and constraints requiring heterogeneous evaluation times and optimal solutions can be analyzed to select a single preferred solution and importantly harnessed to reveal vital design features common to optimal solutions as design principles.
△ Less
Submitted 3 June, 2022;
originally announced June 2022.
-
Counterfactual Explanation of Brain Activity Classifiers using Image-to-Image Transfer by Generative Adversarial Network
Authors:
Teppei Matsui,
Masato Taki,
Trung Quang Pham,
Junichi Chikazoe,
Koji Jimura
Abstract:
Deep neural networks (DNNs) can accurately decode task-related information from brain activations. However, because of the nonlinearity of the DNN, the decisions made by DNNs are hardly interpretable. One of the promising approaches for explaining such a black-box system is counterfactual explanation. In this framework, the behavior of a black-box system is explained by comparing real data and rea…
▽ More
Deep neural networks (DNNs) can accurately decode task-related information from brain activations. However, because of the nonlinearity of the DNN, the decisions made by DNNs are hardly interpretable. One of the promising approaches for explaining such a black-box system is counterfactual explanation. In this framework, the behavior of a black-box system is explained by comparing real data and realistic synthetic data that are specifically generated such that the black-box system outputs an unreal outcome. Here we introduce a novel generative DNN (counterfactual activation generator, CAG) that can provide counterfactual explanations for DNN-based classifiers of brain activations. Importantly, CAG can simultaneously handle image transformation among multiple classes associated with different behavioral tasks. Using CAG, we demonstrated counterfactual explanation of DNN-based classifiers that learned to discriminate brain activations of seven behavioral tasks. Furthermore, by iterative applications of CAG, we were able to enhance and extract subtle spatial brain activity patterns that affected the classifier's decisions. Together, these results demonstrate that the counterfactual explanation based on image-to-image transformation would be a promising approach to understand and extend the current application of DNNs in fMRI analyses.
△ Less
Submitted 28 October, 2021;
originally announced October 2021.
-
Use of statistical outlier detection method in adaptive evolutionary algorithms
Authors:
James M. Whitacre,
Tuan Q. Pham,
Ruhul A. Sarker
Abstract:
In this paper, the issue of adapting probabilities for Evolutionary Algorithm (EA) search operators is revisited. A framework is devised for distinguishing between measurements of performance and the interpretation of those measurements for purposes of adaptation. Several examples of measurements and statistical interpretations are provided. Probability value adaptation is tested using an EA wit…
▽ More
In this paper, the issue of adapting probabilities for Evolutionary Algorithm (EA) search operators is revisited. A framework is devised for distinguishing between measurements of performance and the interpretation of those measurements for purposes of adaptation. Several examples of measurements and statistical interpretations are provided. Probability value adaptation is tested using an EA with 10 search operators against 10 test problems with results indicating that both the type of measurement and its statistical interpretation play significant roles in EA performance. We also find that selecting operators based on the prevalence of outliers rather than on average performance is able to provide considerable improvements to adaptive methods and soundly outperforms the non-adaptive case.
△ Less
Submitted 3 July, 2009;
originally announced July 2009.
-
Credit Assignment in Adaptive Evolutionary Algorithms
Authors:
James M. Whitacre,
Tuan Q. Pham,
Ruhul A. Sarker
Abstract:
In this paper, a new method for assigning credit to search operators is presented. Starting with the principle of optimizing search bias, search operators are selected based on an ability to create solutions that are historically linked to future generations. Using a novel framework for defining performance measurements, distributing credit for performance, and the statistical interpretation of…
▽ More
In this paper, a new method for assigning credit to search operators is presented. Starting with the principle of optimizing search bias, search operators are selected based on an ability to create solutions that are historically linked to future generations. Using a novel framework for defining performance measurements, distributing credit for performance, and the statistical interpretation of this credit, a new adaptive method is developed and shown to outperform a variety of adaptive and non-adaptive competitors.
△ Less
Submitted 3 July, 2009;
originally announced July 2009.
