• Title, Summary, Keyword: multi-task learning

Search Result 60, Processing Time 0.038 seconds

Additional Learning Framework for Multipurpose Image Recognition

  • Itani, Michiaki;Iyatomi, Hitoshi;Hagiwara, Masafumi
    • Proceedings of the Korean Institute of Intelligent Systems Conference
    • /
    • /
    • pp.480-483
    • /
    • 2003
  • We propose a new framework that aims at multi-purpose image recognition, a difficult task for the conventional rule-based systems. This framework is farmed based on the idea of computer-based learning algorithm. In this research, we introduce the new functions of an additional learning and a knowledge reconstruction on the Fuzzy Inference Neural Network (FINN) (1) to enable the system to accommodate new objects and enhance the accuracy as necessary. We examine the capability of the proposed framework using two examples. The first one is the capital letter recognition task from UCI machine learning repository to estimate the effectiveness of the framework itself, Even though the whole training data was not given in advance, the proposed framework operated with a small loss of accuracy by introducing functions of the additional learning and the knowledge reconstruction. The other is the scenery image recognition. We confirmed that the proposed framework could recognize images with high accuracy and accommodate new object recursively.

  • PDF

Direct Learning Control for a Class of Multi-Input Multi-Output Nonlinear Systems (다입력 다출력 비선형시스템에 대한 직접학습제어)

  • 안현식
    • Journal of the Institute of Electronics Engineers of Korea SC
    • /
    • v.40 no.2
    • /
    • pp.19-25
    • /
    • 2003
  • For a class of multi-input multi-output nonlinear systems which perform a given task repetitively, an extended type of a direct leaning control (DLC) is proposed using the information on the (vector) relative degree of a multi-input multi-output system. Existing DLC methods are observed to be applied to a limited class of systems with the relative degree one and a new DLC law is suggested which can be applied to systems having higher relative degree. Using the proposed control law, the control input corresponding to the new desired output trajectory is synthesized directly based on the control inputs obtained from the learning process for other output trajectories. To show the validity and the performance of the proposed DLC, simulations are performed for trajectory tracking control of a two-axis SCARA robot.

The Effect of Worker Heterogeneity in Learning and Forgetting on System Productivity (학습과 망각에 대한 작업자들의 이질성 정도가 시스템 생산성에 미치는 영향)

  • Kim, Sungsu
    • Journal of the Korean Operations Research and Management Science Society
    • /
    • v.40 no.4
    • /
    • pp.145-156
    • /
    • 2015
  • Incorporation of individual learning and forgetting behaviors within worker-task assignment models produces a mixed integer nonlinear program (MINLP) problem, which is difficult to solve as a NP hard due to its nonlinearity in the objective function. Previous studies commonly assume homogeneity among workers in workforce scheduling that takes account of learning and forgetting characteristics. This paper expands previous researches by considering heterogeneous individual learning/forgetting, and investigates the impact of worker heterogeneity in initial expertise, steady-state productivity, learning and forgetting on system performance to assist manager's decision-making in worker-task assignments without tackling complex MINLP models. In order to understand the performance implications of workforce heterogeneity, this paper examines analytically how heterogeneity in each of the four parameters of the exponential learning and forgetting (L/F) model affects system performance in three cases : consecutive assignments with no break, n breaks of s-length each, and total b break-periods occurred over T periods. The study presents the direction of change in worker performance under different assignment schedules as the variance in initial expertise, steady-state productivity, learning or forgetting increases. Thus, it implies whether having more heterogenous workforce in terms of each of four parameters in the L/F model is desired or not in different schedules from the perspective of system productivity measurement.

Research Trends on Deep Reinforcement Learning (심층 강화학습 기술 동향)

  • Jang, S.Y.;Yoon, H.J.;Park, N.S.;Yun, J.K.;Son, Y.S.
    • Electronics and Telecommunications Trends
    • /
    • v.34 no.4
    • /
    • pp.1-14
    • /
    • 2019
  • Recent trends in deep reinforcement learning (DRL) have revealed the considerable improvements to DRL algorithms in terms of performance, learning stability, and computational efficiency. DRL also enables the scenarios that it covers (e.g., partial observability; cooperation, competition, coexistence, and communications among multiple agents; multi-task; decentralized intelligence) to be vastly expanded. These features have cultivated multi-agent reinforcement learning research. DRL is also expanding its applications from robotics to natural language processing and computer vision into a wide array of fields such as finance, healthcare, chemistry, and even art. In this report, we briefly summarize various DRL techniques and research directions.

Multi-modal Sensor System and Database for Human Detection and Activity Learning of Robot in Outdoor (실외에서 로봇의 인간 탐지 및 행위 학습을 위한 멀티모달센서 시스템 및 데이터베이스 구축)

  • Uhm, Taeyoung;Park, Jeong-Woo;Lee, Jong-Deuk;Bae, Gi-Deok;Choi, Young-Ho
    • Journal of Korea Multimedia Society
    • /
    • v.21 no.12
    • /
    • pp.1459-1466
    • /
    • 2018
  • Robots which detect human and recognize action are important factors for human interaction, and many researches have been conducted. Recently, deep learning technology has developed and learning based robot's technology is a major research area. These studies require a database to learn and evaluate for intelligent human perception. In this paper, we propose a multi-modal sensor-based image database condition considering the security task by analyzing the image database to detect the person in the outdoor environment and to recognize the behavior during the running of the robot.

Two person Interaction Recognition Based on Effective Hybrid Learning

  • Ahmed, Minhaz Uddin;Kim, Yeong Hyeon;Kim, Jin Woo;Bashar, Md Rezaul;Rhee, Phill Kyu
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • v.13 no.2
    • /
    • pp.751-770
    • /
    • 2019
  • Action recognition is an essential task in computer vision due to the variety of prospective applications, such as security surveillance, machine learning, and human-computer interaction. The availability of more video data than ever before and the lofty performance of deep convolutional neural networks also make it essential for action recognition in video. Unfortunately, limited crafted video features and the scarcity of benchmark datasets make it challenging to address the multi-person action recognition task in video data. In this work, we propose a deep convolutional neural network-based Effective Hybrid Learning (EHL) framework for two-person interaction classification in video data. Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a state-of-the-art detector for image classification). We broaden a semi-supervised learning method combined with an active learning method to improve overall performance. Numerous types of two-person interactions exist in the real world, which makes this a challenging task. In our experiment, we consider a limited number of actions, such as hugging, fighting, linking arms, talking, and kidnapping in two environment such simple and complex. We show that our trained model with an active semi-supervised learning architecture gradually improves the performance. In a simple environment using an Intelligent Technology Laboratory (ITLab) dataset from Inha University, performance increased to 95.6% accuracy, and in a complex environment, performance reached 81% accuracy. Our method reduces data-labeling time, compared to supervised learning methods, for the ITLab dataset. We also conduct extensive experiment on Human Action Recognition benchmarks such as UT-Interaction dataset, HMDB51 dataset and obtain better performance than state-of-the-art approaches.

Multi-Scale, Multi-Object and Real-Time Face Detection and Head Pose Estimation Using Deep Neural Networks (다중크기와 다중객체의 실시간 얼굴 검출과 머리 자세 추정을 위한 심층 신경망)

  • Ahn, Byungtae;Choi, Dong-Geol;Kweon, In So
    • The Journal of Korea Robotics Society
    • /
    • v.12 no.3
    • /
    • pp.313-321
    • /
    • 2017
  • One of the most frequently performed tasks in human-robot interaction (HRI), intelligent vehicles, and security systems is face related applications such as face recognition, facial expression recognition, driver state monitoring, and gaze estimation. In these applications, accurate head pose estimation is an important issue. However, conventional methods have been lacking in accuracy, robustness or processing speed in practical use. In this paper, we propose a novel method for estimating head pose with a monocular camera. The proposed algorithm is based on a deep neural network for multi-task learning using a small grayscale image. This network jointly detects multi-view faces and estimates head pose in hard environmental conditions such as illumination change and large pose change. The proposed framework quantitatively and qualitatively outperforms the state-of-the-art method with an average head pose mean error of less than $4.5^{\circ}$ in real-time.

A biologically inspired model based on a multi-scale spatial representation for goal-directed navigation

  • Li, Weilong;Wu, Dewei;Du, Jia;Zhou, Yang
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • v.11 no.3
    • /
    • pp.1477-1491
    • /
    • 2017
  • Inspired by the multi-scale nature of hippocampal place cells, a biologically inspired model based on a multi-scale spatial representation for goal-directed navigation is proposed in order to achieve robotic spatial cognition and autonomous navigation. First, a map of the place cells is constructed in different scales, which is used for encoding the spatial environment. Then, the firing rate of the place cells in each layer is calculated by the Gaussian function as the input of the Q-learning process. The robot decides on its next direction for movement through several candidate actions according to the rules of action selection. After several training trials, the robot can accumulate experiential knowledge and thus learn an appropriate navigation policy to find its goal. The results in simulation show that, in contrast to the other two methods(G-Q, S-Q), the multi-scale model presented in this paper is not only in line with the multi-scale nature of place cells, but also has a faster learning potential to find the optimized path to the goal. Additionally, this method also has a good ability to complete the goal-directed navigation task in large space and in the environments with obstacles.

A Case Study on the Instructional Dimensions in Teaching Mathematics to the Elementary School Student from Multi-cultural Backgrounds (다문화권 학생들의 초등수학 학습과정에 관한 사례연구)

  • Jang, Yun-Young;ChoiKoh, Sang-Sook
    • The Mathematical Education
    • /
    • v.48 no.4
    • /
    • pp.419-442
    • /
    • 2009
  • This study was to find the difficulties students faced in their mathematical learning and to identify the instructional dimensions a teacher provided for the students from multi-cultural background. Since the study was focused on the process of students' learning, the qualitative method was chosen through clinical interviews with 2 students in a total of 11 units which played a role of compensating their learning of mathematics as an extra curriculum. The students solved the computational problems relying on formal procedure without understanding of concepts and principles and solved the word problems based on own interpretation of certain words without semantic comprehension out of math sentences. As the instructional dimensions of teaching mathematics, tasks, a tool and classroom norm were found in the activities they performed. For the tasks, situated tasks, challenging tasks, tasks with lack of conditions, and open-ended exploratory tasks were used. As the tool, pictorial representations were very useful to describe their ideas. Finally, as the classroom norm, consider equity for everyone, and cooperate and encourage each other were found.

  • PDF

Application of Multi-agent Reinforcement Learning to CELSS Material Circulation Control

  • Hirosaki, Tomofumi;Yamauchi, Nao;Yoshida, Hiroaki;Ishikawa, Yoshio;Miyajima, Hiroyuki
    • Proceedings of the Korea Inteligent Information System Society Conference
    • /
    • /
    • pp.145-150
    • /
    • 2001
  • A Controlled Ecological Life Support System(CELSS) is essential for man to live a long time in a closed space such as a lunar base or a mars base. Such a system may be an extremely complex system that has a lot of facilities and circulates multiple substances,. Therefore, it is very difficult task to control the whole CELSS. Thus by regarding facilities constituting the CELSS as agents and regarding the status and action as information, the whole CELSS can be treated as multi-agent system(MAS). If a CELSS can be regarded as MAS the CELSS can have three advantages with the MAS. First the MAS need not have a central computer. Second the expendability of the CELSS increases. Third, its fault tolerance rises. However it is difficult to describe the cooperation protocol among agents for MAS. Therefore in this study we propose to apply reinforcement learning (RL), because RL enables and agent to acquire a control rule automatically. To prove that MAS and RL are effective methods. we have created the system in Java, which easily gives a distributed environment that is the characteristics feature of an agent. In this paper, we report the simulation results for material circulation control of the CELSS by the MAS and RL.

  • PDF