• The KIPS Transactions:PartB
• /
• v.8B no.6
• /
• pp.601-608
• /
• 2001

The automated assignment of multiple tasks to executing agents is a key problem in the area of multi-agent systems. In many domains, significant savings can be achieved by reallocating tasks among agents with different costs for handling tasks. The automation of task reallocation among self-interested agents requires that the individual agents use a common negotiation protocol that prescribes how they have to interact in order to come to an agreement on "who does what". In this paper, we introduce the multi-agent Traveling Salesman Problem(TSP) as an example of task reallocation problem, and suggest the Vickery auction as an interagent negotiation protocol for solving this problem. In general, auction-based protocols show several advantageous features: they are easily implementable, they enforce an efficient assignment process, and they guarantce an agreement even in scenarios in which the agents possess only very little domain-specific Knowledge. Furthermore Vickrey auctions have the additional advantage that each interested agent bids only once and that the dominant strategy is to bid one′s true valuation. In order to apply this market-based protocol into task reallocation among self-interested agents, we define the profit of each agent, the goal of negotiation, tasks to be traded out through auctions, the bidding strategy, and the sequence of auctions. Through several experiments with sample multi-agent TSPs, we show that the task allocation can improve monotonically at each step and then finally an optimal task allocation can be found with this protocol.

• Journal of the Korean Society of Radiology
• /
• v.83 no.6
• /
• pp.1208-1218
• /
• 2022

In the medical field, artificial intelligence has been used in various ways with many developments. However, most artificial intelligence technologies are developed so that one model can perform only one task, which is a limitation in designing the complex reading process of doctors with artificial intelligence. Multi-task learning is an optimal way to overcome the limitations of single-task learning methods. Multi-task learning can create a model that is efficient and advantageous for generalization by simultaneously integrating various tasks into one model. This study investigated the concepts, types, and similar concepts as multi-task learning, and examined the status and future possibilities of multi-task learning in the medical research.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.4
• /
• pp.1100-1122
• /
• 2023

Scheduling user-submitted cloud tasks to the appropriate virtual machine (VM) in cloud computing is critical for cloud providers. However, as the demand for cloud resources from user tasks continues to grow, current evolutionary algorithms (EAs) cannot satisfy the optimal solution of large-scale cloud task scheduling problems. In this paper, we first construct a large- scale multi-objective cloud task problem considering the time and cost functions. Second, a multi-objective optimization algorithm based on multi-factor optimization (MFO) is proposed to solve the established problem. This algorithm solves by decomposing the large-scale optimization problem into multiple optimization subproblems. This reduces the computational burden of the algorithm. Later, the introduction of the MFO strategy provides the algorithm with a parallel evolutionary paradigm for multiple subpopulations of implicit knowledge transfer. Finally, simulation experiments and comparisons are performed on a large-scale task scheduling test set on the CloudSim platform. Experimental results show that our algorithm can obtain the best scheduling solution while maintaining good results of the objective function compared with other optimization algorithms.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.11
• /
• pp.5546-5559
• /
• 2019

Facial action units (AUs) have recently drawn increased attention because they can be used to recognize facial expressions. A variety of methods have been designed for frontal-view AU detection, but few have been able to handle multi-view face images. In this paper we propose a method for multi-view facial AU detection using a fused multilayer, multi-task, and multi-label deep learning network. The network can complete two tasks: AU detection and facial view detection. AU detection is a multi-label problem and facial view detection is a single-label problem. A residual network and multilayer fusion are applied to obtain more representative features. Our method is effective and performs well. The F1 score on FERA 2017 is 13.1% higher than the baseline. The facial view recognition accuracy is 0.991. This shows that our multi-task, multi-label model could achieve good performance on the two tasks.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.308-310
• /
• 2022

With the rise of the Al wave, task-oriented dialogue systems have become one of the popular research directions in academia and industry. Currently, task-oriented dialogue systems mainly adopt pipelined form, which mainly includes natural language understanding, dialogue state decision making, dialogue state tracking and natural language generation. However, pipelining is prone to error propagation, so many task-oriented dialogue systems in the market are only for single-round dialogues. Usually single- domain dialogues have relatively accurate semantic understanding, while they tend to perform poorly on multi-domain, multi-round dialogue datasets. To solve these issues, we developed a paddle-based multi-domain task-oriented Chinese dialogue system. It is based on NEZHA-base pre-training model and CrossWOZ dataset, and uses intention recognition module, dichotomous slot recognition module and NER recognition module to do DST and generate replies based on rules. Experiments show that the dialogue system not only makes good use of the context, but also effectively addresses long-term dependencies. In our approach, the DST of dialogue tracking state is improved, and our DST can identify multiple slotted key-value pairs involved in the discourse, which eliminates the need for manual tagging and thus greatly saves manpower.

• Journal of Broadcast Engineering
• /
• v.25 no.5
• /
• pp.742-749
• /
• 2020

This paper proposes a method for recognizing both phoneme and emotion using a common neural network and a multi-task training method for the common neural network. The common neural network performs the same function for both recognition tasks, which corresponds to the structure of multi-information recognition of human using a single auditory system. The multi-task training conducts a feature modeling that is commonly applicable to multiple information and provides generalized training, which enables to improve the performance by reducing an overfitting occurred in the conventional individual training for each information. A method for increasing phoneme recognition performance is also proposed that applies weight to the phoneme in the multi-task training. When using the same feature vector and neural network, it is confirmed that the proposed common neural network with multi-task training provides higher performance than the individual one trained for each task.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.383-392
• /
• 2023

Post-earthquake building condition assessment is crucial for subsequent rescue and remediation and can be automated by emerging computer vision and deep learning technologies. This study is based on an endeavour for the 2nd International Competition of Structural Health Monitoring (IC-SHM 2021). The task package includes five image segmentation objectives - defects (crack/spall/rebar exposure), structural component, and damage state. The structural component and damage state tasks are identified as the priority that can form actionable decisions. A multi-task Convolutional Neural Network (CNN) is proposed to conduct the two major tasks simultaneously. The rest 3 sub-tasks (spall/crack/rebar exposure) were incorporated as auxiliary tasks. By synchronously learning defect information (spall/crack/rebar exposure), the multi-task CNN model outperforms the counterpart single-task models in recognizing structural components and estimating damage states. Particularly, the pixel-level damage state estimation witnesses a mIoU (mean intersection over union) improvement from 0.5855 to 0.6374. For the defect detection tasks, rebar exposure is omitted due to the extremely biased sample distribution. The segmentations of crack and spall are automated by single-task U-Net but with extra efforts to resample the provided data. The segmentation of small objects (spall and crack) benefits from the resampling method, with a substantial IoU increment of nearly 10%.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.8
• /
• pp.319-330
• /
• 2021

Recently, in the field of open domain natural language question answering, multi-task, multi-hop question answering has been studied extensively. In this paper, we propose a novel deep neural network model using hierarchical graphs to answer effectively such multi-task, multi-hop questions. The proposed model extracts different levels of contextual information from multiple paragraphs using hierarchical graphs and graph neural networks, and then utilize them to predict answer type, supporting sentences and answer spans simultaneously. Conducting experiments with the HotpotQA benchmark dataset, we show high performance and positive effects of the proposed model.

• Korean Journal of Child Studies
• /
• v.38 no.3
• /
• pp.75-90
• /
• 2017

Objective: This study aimed to identify children's cognitive processing and performance characteristics while multiple task performance. It confirmed whether their multiple task performance and psychological refractory period (PRP) varied by task condition (stimulus onset asynchrony [SOA] and task difficulty) and stimulus modality. Methods: Seventy 5-year-olds were recruited. Multi-task tools were developed using the E-prime software. The children were required to respond to two stimuli (visual or auditory) presented with microscopic time difference and their response times (RTs) were recorded. Results: As the SOA increased, the RTs in the first task increased, while the RTs in the second task and PRP decreased. The RTs of the first and second tasks, and the PRP for difficult tasks, were significantly longer than those for easy tasks were. Additionally, there was an interaction effect between the SOA and task difficulty. Although there was no main effect of stimulus modality, task difficulty moderated the modality effect. In the high difficulty condition, the RTs of the first and second tasks and PRP for the visual-visual task were significantly longer than those for auditory-auditory task were. Conclusion: These results inform theoretical discussions on children's multi-task mechanism, and the loss of multiple task performance. Additionally, they provide practical implications and information on the composition of multi-tasks suitable for children in educational environments.

• International Journal of Computer Science & Network Security
• /
• v.23 no.3
• /
• pp.31-48
• /
• 2023

The popularization of smart devices and subsequent optimization of their sensing capacity has resulted in a novel mobile crowdsensing (MCS) pattern, which employs smart devices as sensing nodes by recruiting users to develop a sensing network for multiple-task performance. This technique has garnered much scholarly interest in terms of sensing range, cost, and integration. The MCS is prevalent in various fields, including environmental monitoring, noise monitoring, and road monitoring. A complete MCS life cycle entails task allocation, data collection, and data aggregation. Regardless, specific drawbacks remain unresolved in this study despite extensive research on this life cycle. This article mainly summarizes single-task, multi-task allocation, and space-time multi-task allocation at the task allocation stage. Meanwhile, the quality, safety, and efficiency of data collection are discussed at the data collection stage. Edge computing, which provides a novel development idea to derive data from the MCS system, is also highlighted. Furthermore, data aggregation security and quality are summarized at the data aggregation stage. The novel development of multi-modal data aggregation is also outlined following the diversity of data obtained from MCS. Overall, this article summarizes the three aspects of the MCS life cycle, analyzes the issues underlying this study, and offers developmental directions for future scholars' reference.