• ETRI Journal
• /
• v.43 no.6
• /
• pp.1004-1012
• /
• 2021

The collaboration productively interacting between multi-agents has become an emerging issue in real-world applications. In reinforcement learning, multi-agent environments present challenges beyond tractable issues in single-agent settings. This collaborative environment has the following highly complex attributes: sparse rewards for task completion, limited communications between each other, and only partial observations. In particular, adjustments in an agent's action policy result in a nonstationary environment from the other agent's perspective, which causes high variance in the learned policies and prevents the direct use of reinforcement learning approaches. Unexpected social loafing caused by high dispersion makes it difficult for all agents to succeed in collaborative tasks. Therefore, we address a paradox caused by the social loafing to significantly reduce total returns after a certain timestep of multi-agent reinforcement learning. We further demonstrate that the collaborative paradox in multi-agent environments can be avoided by our proposed effective early stop method leveraging a metric for social loafing.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.5
• /
• pp.593-600
• /
• 2019

The similarity in tones between speakers can lower the performance of speaker verification. To improve the performance of speaker verification systems, we propose a multi-task learning technique using deep neural network to learn speaker information and age information. Multi-task learning can improve generalization performances, because it helps deep neural networks to prevent hidden layers from overfitting into one task. However, we found in experiments that learning of age information does not work well in the process of learning the deep neural network. In order to improve the learning, we propose a method to dynamically change the objective function weights of speaker identification and age estimation in the learning process. Results show the equal error rate based on RSR2015 evaluation data set, 6.91 % for the speaker verification system without using age information, 6.77 % using age information only, and 4.73 % using age information when weight change technique was applied.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.9
• /
• pp.7-13
• /
• 2018

Recent advancement in data gathering technique improves the capability of information collecting, thus allowing the learning process between gathered data patterns and application sub-tasks. A pattern can be associated with multiple labels, demanding multi-label learning capability, resulting in significant attention to multi-label feature selection since it can improve multi-label learning accuracy. However, existing evolutionary multi-label feature selection methods suffer from ineffective search process. In this study, we propose a evolutionary search process for the task of multi-label feature selection problem. The proposed method creates large set of offspring or new feature subsets and then retains the most promising feature subset. Experimental results demonstrate that the proposed method can identify feature subsets giving good multi-label classification accuracy much faster than conventional methods.

• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.351-360
• /
• 2018

Human emotion recognition is a research topic that is receiving continuous attention in computer vision and artificial intelligence domains. This paper proposes a method for classifying human emotions through multiple neural networks based on multi-modal signals which consist of image, landmark, and audio in a wild environment. The proposed method has the following features. First, the learning performance of the image-based network is greatly improved by employing both multi-task learning and semi-supervised learning using the spatio-temporal characteristic of videos. Second, a model for converting 1-dimensional (1D) landmark information of face into two-dimensional (2D) images, is newly proposed, and a CNN-LSTM network based on the model is proposed for better emotion recognition. Third, based on an observation that audio signals are often very effective for specific emotions, we propose an audio deep learning mechanism robust to the specific emotions. Finally, so-called emotion adaptive fusion is applied to enable synergy of multiple networks. The proposed network improves emotion classification performance by appropriately integrating existing supervised learning and semi-supervised learning networks. In the fifth attempt on the given test set in the EmotiW2017 challenge, the proposed method achieved a classification accuracy of 57.12%.

• Journal of Multimedia Information System
• /
• v.9 no.2
• /
• pp.75-86
• /
• 2022

The existing image quality assessment (IQA) datasets have a small number of samples. Some methods based on transfer learning or data augmentation cannot make good use of image quality-related features. A No Reference (NR)-IQA method based on multi-task training and quality awareness is proposed. First, single or multiple distortion types and levels are imposed on the original image, and different strategies are used to augment different types of distortion datasets. With the idea of weak supervision, we use the Full Reference (FR)-IQA methods to obtain the pseudo-score label of the generated image. Then, we combine the classification information of the distortion type, level, and the information of the image quality score. The ResNet50 network is trained in the pre-train stage on the augmented dataset to obtain more quality-aware pre-training weights. Finally, the fine-tuning stage training is performed on the target IQA dataset using the quality-aware weights to predicate the final prediction score. Various experiments designed on the synthetic distortions and authentic distortions datasets (LIVE, CSIQ, TID2013, LIVEC, KonIQ-10K) prove that the proposed method can utilize the image quality-related features better than the method using only single-task training. The extracted quality-aware features improve the accuracy of the model.

• English Language & Literature Teaching
• /
• v.11 no.4
• /
• pp.365-389
• /
• 2005

The purpose of this study is to present the task model that is the most effective in English language methodology based on the investigation of task-based performance in Korean EFL classroom practice. The subjects were 538 high school students and 126 high school teachers, each of whom had common experiences using the materials of task-based activities for more than one year. To analyze the data, the program SPSS WIN 11.0 including frequency distribution and chi-square analysis was used. The results of the questionnaire analysis showed that both teachers and students had a comparatively high level of satisfaction in task rationale, but that they had some mixed responses in the fields of input data, settings, and activity types. To conclude, a few suggestions are made to provide some meaningful considerations for the EFL teachers and material developers: a) task goals and rationale that encourage the learner's positive motivation; b) authenticity of input data based on the real-world context; c) collaborative learning environment that enhances communicative interaction; d) proportional representation of the creative problem-solving activities related to discussions and decision-making processes; e) systematic introduction of integrated language skills. It also suggests that the multi-lateral task model, which has some positive assets compared to previous task models, be newly introduced and applied to the second language learning classrooms.

• The Journal of Korea Robotics Society
• /
• v.8 no.3
• /
• pp.217-227
• /
• 2013

We introduce a learning system for the sight reading of simple drum sequences. Sight reading is a cognitive-motor skill that requires reading of music symbols and actions of multiple limbs for playing the music. The system provides knowledge of results (KR) pertaining to the learner's performance by color-coding music symbols, and guides the learner by indicating the corresponding action for a given music symbol using additional auditory or vibrotactile cues. To evaluate the effects of KR and guidance cues, three learning methods were experimentally compared: KR only, KR with auditory cues, and KR with vibrotactile cues. The task was to play a random 16-note-long drum sequence displayed on a screen. Thirty university students learned the task using one of the learning methods in a between-subjects design. The experimental results did not show statistically significant differences between the methods in terms of task accuracy and completion time.

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.192-198
• /
• 2024

This paper investigates the application of multi-agent deep reinforcement learning in the fighting game Samurai Shodown using Proximal Policy Optimization (PPO) and Advantage Actor-Critic (A2C) algorithms. Initially, agents are trained separately for 200,000 timesteps using Convolutional Neural Network (CNN) and Multi-Layer Perceptron (MLP) with LSTM networks. PPO demonstrates superior performance early on with stable policy updates, while A2C shows better adaptation and higher rewards over extended training periods, culminating in A2C outperforming PPO after 1,000,000 timesteps. These findings highlight PPO's effectiveness for short-term training and A2C's advantages in long-term learning scenarios, emphasizing the importance of algorithm selection based on training duration and task complexity. The code can be found in this link https://github.com/Lexer04/Samurai-Shodown-with-Reinforcement-Learning-PPO.

• The Journal of Korea Robotics Society
• /
• v.18 no.2
• /
• pp.225-232
• /
• 2023

With the recent increase of multiple robots cooperating in smart manufacturing logistics environments, it has become very important how to predict the safety and efficiency of the individual tasks and dynamically assign them to the best one of available robots. In this paper, we propose a novel task policy learner based on deep relational reinforcement learning for predicting the safety and efficiency of tasks in a multi-robot manufacturing logistics environment. To reduce learning complexity, the proposed system divides the entire safety/efficiency prediction process into two distinct steps: the policy parameter estimation and the rule-based policy inference. It also makes full use of domain-specific knowledge for policy rule learning. Through experiments conducted with virtual dynamic manufacturing logistics environments using NVIDIA's Isaac simulator, we show the effectiveness and superiority of the proposed system.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.483-486
• /
• 2020

Age estimation and gender classification for human is a classic problem in computer vision. Almost research focus just only one task and the models are too heavy to run on low-cost system. In our research, we aim to apply multitasking learning to perform both task on a lightweight model which can achieve good precision on embedded system in the real time.