Intisar Chowdhury
Intisar Chowdhury has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Qiangfu Zhao
System Intelligence Laboratory, The University of Aizu, Aizu-Wakamatsu, Japan
3 shared publications - Kai Su
Graduate School of Computer Science and Engineering, The University of Aizu, Aizu-Wakamatsu 965-8580, Japan
2 shared publications - Alexander Vazhenin
University of Aizu
1 shared publication Ruth Cortez
Department of Computer Science and Engineering, University of Aizu
1 shared publication- Yong Liu
Graduate School of Computer Science and Engineering, The University of Aizu, Aizu-Wakamatsu 965-8580, Japan
1 shared publication - Yutaka Watanobe
Department of Computer Science and Engineering, The University of Aizu, Aizuwakamatsu 965-8580, Fukushima, Japan
1 shared publication
Latest Publications
- Open Access
- 30 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 93871-93891
CMNN: Coupled Modular Neural Network
In this paper, we propose a multi-branch neural network architecture named Coupled Modular Neural Network (CMNN). A CMNN is a network consisting of$\beta $ closely coupled sub-networks, where$\beta $ is termed as the branching factor in this paper. We call the whole network a super-graph and each sub-network a sub-graph. Each sub-graph is a stand-alone neural network and shares a common block with other sub-graphs. To effectively leverage the super-graph we propose a simple but easy-to-implement Round-Robin-based learning algorithm. Each training iteration contains two phases. In the first phase, we choose a sub-graph in a Round-Robin fashion and train it using knowledge of the super-graph (distillation). In the second phase, we fine-tune the super-graph based on the updated sub-graphs. This algorithm produces a different copy of the super-graph at each iteration which acts as an improved teacher network for the sub-graph; and a different copy of one of the sub-graphs which functions as a new building block for the super-graph. To validate and test CMNN and the proposed algorithm, we conduct experiments on CIFAR-10, CIFAR-100, Tiny ImageNet and a private On-Road-Risk (ORR) datasets. Empirical results on all these four datasets indicate that we not only obtain a strong sub-graph network, the learning framework can also produce strong ensemble performance which substantiates the diversity introduced throughout the learning framework.- 30 June 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 93871-93891
Open AccessCMNN: Coupled Modular Neural Network
In this paper, we propose a multi-branch neural network architecture named Coupled Modular Neural Network (CMNN). A CMNN is a network consisting of$\beta $ closely coupled sub-networks, where$\beta $ is termed as the branching factor in this paper. We call the whole network a super-graph and each sub-network a sub-graph. Each sub-graph is a stand-alone neural network and shares a common block with other sub-graphs. To effectively leverage the super-graph we propose a simple but easy-to-implement Round-Robin-based learning algorithm. Each training iteration contains two phases. In the first phase, we choose a sub-graph in a Round-Robin fashion and train it using knowledge of the super-graph (distillation). In the second phase, we fine-tune the super-graph based on the updated sub-graphs. This algorithm produces a different copy of the super-graph at each iteration which acts as an improved teacher network for the sub-graph; and a different copy of one of the sub-graphs which functions as a new building block for the super-graph. To validate and test CMNN and the proposed algorithm, we conduct experiments on CIFAR-10, CIFAR-100, Tiny ImageNet and a private On-Road-Risk (ORR) datasets. Empirical results on all these four datasets indicate that we not only obtain a strong sub-graph network, the learning framework can also produce strong ensemble performance which substantiates the diversity introduced throughout the learning framework. - Open Access
- 30 September 2020
- Published by Springer Nature in International Journal of Machine Learning and Cybernetics
- Vol. 12 (3) , 763-781
MS-NET: modular selective network
We propose a modular architecture of Deep Neural Network (DNN) for multi-class classification task. The architecture consists of two parts, a router network and a set of expert networks. In this architecture, for a C-class classification problem, we have exactly C experts. The backbone network for these experts and the router are built with simple and identical DNN architecture. For each class, the modular network has a certain number $$\rho$$ of expert networks specializing in that particular class, where $$\rho$$ is called the redundancy rate in this study. We demonstrate that $$\rho$$ plays a vital role in the performance of the network. Although these experts are light weight and weak learners alone, together they match the performance of more complex DNNs. We train the network in two phase wherein, first the router is trained on the whole set of training data followed by training each expert network enforced by a new stochastic objective function that facilitates alternative training on a small subset of expert data and the whole set of data. This alternative training provides an additional form of regularization and avoids over-fitting the expert network on subset data. During the testing phase, the router dynamically selects a fixed number of experts for further evaluation of the input datum. The modular nature and low parameter requirement of the network makes it very suitable in distributed and low computational environments. Extensive empirical study and theoretical analysis on CIFAR-10, CIFAR-100 and F-MNIST substantiate the effectiveness and efficiency of our proposed modular network.- 30 September 2020
- Published by Springer Nature in International Journal of Machine Learning and Cybernetics
- Vol. 12 (3) , 763-781
Open AccessMS-NET: modular selective network
We propose a modular architecture of Deep Neural Network (DNN) for multi-class classification task. The architecture consists of two parts, a router network and a set of expert networks. In this architecture, for a C-class classification problem, we have exactly C experts. The backbone network for these experts and the router are built with simple and identical DNN architecture. For each class, the modular network has a certain number $$\rho$$ of expert networks specializing in that particular class, where $$\rho$$ is called the redundancy rate in this study. We demonstrate that $$\rho$$ plays a vital role in the performance of the network. Although these experts are light weight and weak learners alone, together they match the performance of more complex DNNs. We train the network in two phase wherein, first the router is trained on the whole set of training data followed by training each expert network enforced by a new stochastic objective function that facilitates alternative training on a small subset of expert data and the whole set of data. This alternative training provides an additional form of regularization and avoids over-fitting the expert network on subset data. During the testing phase, the router dynamically selects a fixed number of experts for further evaluation of the input datum. The modular nature and low parameter requirement of the network makes it very suitable in distributed and low computational environments. Extensive empirical study and theoretical analysis on CIFAR-10, CIFAR-100 and F-MNIST substantiate the effectiveness and efficiency of our proposed modular network. - Open Access
- 8 May 2020
- Published by EDP Sciences in SHS Web of Conferences
- Vol. 77, 01004
Next-Generation Programming Learning Platform: Architecture and Challenges
With the rapid development of information technology, programming has become a vital skill. An online judge system can be used as a programming education platform, where the daily activities of users and judges are used to generate useful learning objects (e.g., tasks, solution codes, evaluations). Intelligent software agents can utilize such objects to create an ecosystem. To implement such an ecosystem, a generic architecture that covers the whole lifecycle of data on the platform and the functionalities of an e-learning system should take into account the particularities of the online judge system. In this paper, an architecture that implements such an ecosystem based on an online judge system is proposed. The potential benefits and research challenges are discussed.
- 8 May 2020
- Published by EDP Sciences in SHS Web of Conferences
- Vol. 77, 01004
Open AccessNext-Generation Programming Learning Platform: Architecture and Challenges
With the rapid development of information technology, programming has become a vital skill. An online judge system can be used as a programming education platform, where the daily activities of users and judges are used to generate useful learning objects (e.g., tasks, solution codes, evaluations). Intelligent software agents can utilize such objects to create an ecosystem. To implement such an ecosystem, a generic architecture that covers the whole lifecycle of data on the platform and the functionalities of an e-learning system should take into account the particularities of the online judge system. In this paper, an architecture that implements such an ecosystem based on an online judge system is proposed. The potential benefits and research challenges are discussed.