Publications
Detailed Information
Task-Oriented Design through Deep Reinforcement Learning
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 곽노준 | - |
dc.contributor.author | 최준영 | - |
dc.date.accessioned | 2019-05-07T04:13:54Z | - |
dc.date.available | 2019-05-07T04:13:54Z | - |
dc.date.issued | 2019-02 | - |
dc.identifier.other | 000000154948 | - |
dc.identifier.uri | https://hdl.handle.net/10371/151422 | - |
dc.description | 학위논문 (석사)-- 서울대학교 대학원 : 융합과학기술대학원 지능형융합시스템학과, 2019. 2. 곽노준. | - |
dc.description.abstract | We propose a new low-cost machine-learning-based methodology which
assists designers in reducing the gap between the problem and the solution in the design process. Our work applies reinforcement learning (RL) to find the optimal task-oriented design solution through the construction of the design action for each task. For this task-oriented design, the 3D design process in product design is assigned to an action space in Deep RL, and a desired 3D model is obtained by training each design action according to the task. By showing that this method achieves satisfactory design even when applied to a task pursuing multiple goals, we suggest the direction of how machine learning can contribute in design process. Also, we have validated with product designers that this methodology can assist the creative part in the process of design. | - |
dc.description.tableofcontents | 1 Introduction 1
2 RelatedWorks 4 2.1 Constructive Design . . . . . . . . . . . . . . . . . . . . . . 4 2.2 Reinforcement Learning . . . . . . . . . . . . . . . . . . . 4 3 Environment 7 3.1 Task Specification . . . . . . . . . . . . . . . . . . . . . . . 7 3.2 3D Simulation Environment . . . . . . . . . . . . . . . . . 8 3.3 Reinforcement Learning Environment . . . . . . . . . . . . 9 4 Experiment 12 4.1 Pouring Environment . . . . . . . . . . . . . . . . . . . . . 12 4.1.1 Quantitative Analysis . . . . . . . . . . . . . . . . . 13 4.1.2 Qualitative Analysis . . . . . . . . . . . . . . . . . 14 4.2 Shaking Environment . . . . . . . . . . . . . . . . . . . . . 15 4.2.1 Quantitative Analysis . . . . . . . . . . . . . . . . . 16 4.2.2 Qualitative Analysis . . . . . . . . . . . . . . . . . 17 4.3 Hybrid-Learning . . . . . . . . . . . . . . . . . . . . . . . 18 4.3.1 Quantitative Analysis . . . . . . . . . . . . . . . . . 19 4.3.2 Qualitative Analysis . . . . . . . . . . . . . . . . . 20 4.4 Contribution in Design Process . . . . . . . . . . . . . . . . 20 ii 5 Conclusion 22 5.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Bibliography 24 Appendix 26 Abstract in Korean 29 | - |
dc.language.iso | eng | - |
dc.publisher | 서울대학교 대학원 | - |
dc.subject.ddc | 620.82 | - |
dc.title | Task-Oriented Design through Deep Reinforcement Learning | - |
dc.type | Thesis | - |
dc.type | Dissertation | - |
dc.description.degree | Master | - |
dc.contributor.affiliation | 융합과학기술대학원 지능형융합시스템학과 | - |
dc.date.awarded | 2019-02 | - |
dc.identifier.uci | I804:11032-000000154948 | - |
dc.identifier.holdings | 000000000026▲000000000039▲000000154948▲ | - |
- Appears in Collections:
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.