Generative Design by Reinforcement Learning: Enhancing the Diversity of Topology Optimization Designs

Jang, Seowoo; Yoo, Soyoung; Kang, Namwoo

doi:10.1016/j.cad.2022.103225

Cited by 47 publications

(17 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reinforcement learning, combined with a simulated annealing algorithm and PSO, can improve the efficiency of searching for the optimal solution. By describing the generative design as a sequence problem and then using the given reference design to find the optimal combination of design parameters, RL is applied to the structural optimisation design of automobile wheels [13]. Q‐learning of RL can be used for the design function of ship optimisation design to simulate the process of human designers using their experience for optimisation design [14].…”

Section: Literature Reviewmentioning

confidence: 99%

Engineering design optimisation using reinforcement learning with episodic controllers

Yang

Cheng

Xiao

et al. 2022

Cognitive Comp and Systems

View full text Add to dashboard Cite

Engineers solving engineering design problems can be regarded as a gradual optimisation process that involves strategising. The process can be modelled as a reinforcement learning (RL) framework. This article presents an RL model with episodic controllers to solve engineering problems. Episodic controllers provide a mechanism for using the short‐term and long‐term memories to improve the efficiency of searching for engineering problem solutions. This work demonstrates that the two kinds of models of memories can be incorporated into the existing RL framework. Finally, an optimised design problem of a crane girder is illustrated by RL with episodic controllers. The work presented in this study leverages the RL model that has been shown to mimic human problem solving in engineering optimised design problems.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Engineering design optimisation using reinforcement learning with episodic controllers

Yang

Cheng

Xiao

et al. 2022

Cognitive Comp and Systems

View full text Add to dashboard Cite

show abstract

“…We use the state-of-the-art GAN in single class image generation called StyleGAN2 [19,20,21], which is capable of generating realistic and high quality images. Further, the research in the design community has proven the efficacy of GAN based automated design synthesis in recent years [15,14,13], which makes GANs a suitable option for datadriven automated design synthesis.…”

Section: Generative Adversarial Networkmentioning

confidence: 99%

“…To overcome this problem, data-driven methods such as generative adversarial networks (GANs) [7] and variational autoencoders (VAEs) [8], have been employed in many design synthesis problems [9,10,11,12,13,14,15]. GANs and VAEs are generally capable of learning complex distributions of existing designs and even considering performance and quality evaluation when generating new designs [9,10].…”

Section: Introductionmentioning

confidence: 99%

CreativeGAN: Editing Generative Adversarial Networks for Creative Design Synthesis

Nobari¹,

Rashad²,

Ahmed³

2021

Preprint

View full text Add to dashboard Cite

Modern machine learning techniques, such as deep neural networks, are transforming many disciplines ranging from image recognition to language understanding, by uncovering patterns in big data and making accurate predictions. They have also shown promising results for synthesizing new designs, which is crucial for creating products and enabling innovation. Generative models, including generative adversarial networks (GANs), have proven to be effective for design synthesis with applications ranging from product design to metamaterial design. These automated computational design methods can support human designers, who typically create designs by a time-consuming process of iteratively exploring ideas using experience and heuristics. However, there are still challenges remaining in automatically synthesizing 'creative' designs. GAN models, however, are not capable of generating unique designs, a key to innovation and a major gap in AI-based design automation applications. This paper proposes an automated method, named CreativeGAN, for generating novel designs. It does so by identifying components that make a design unique and modifying a GAN model such that it becomes more likely to generate designs with identified unique components. The method combines state-of-art novelty detection, segmentation, novelty localization, rewriting, and generative models for creative design synthesis. Using a dataset of bicycle designs, we demonstrate that the method can create new bicycle designs with unique frames and handles, and generalize rare novelties * Address all correspondence to this author. to a broad set of designs. Our automated method requires no human intervention and demonstrates a way to rethink creative design synthesis and exploration.

show abstract

“…Parametric design is a method of parametric model creation based on particular variables: dimensions, quantities, or geometries [8]. Generative design is a design exploration method to produce optimum designs by employing topology optimization while facing several limitations [9]. Algorithmic design is a design procedure that utilizes algorithms: Visual Programming Language (VPL) and Textual Programming Language (TPL) [10].…”

Section: Introductionmentioning

confidence: 99%

Bibliometric Analysis of Generative Design, Algorithmic Design, and Parametric Design in Architecture

Michelle

Gemilang²

2022

JARINA

View full text Add to dashboard Cite

This research aims to display, compare, and analyze the keywords related to parametric design, generative design, and algorithmic design. Digital design has become increasingly inseparable from architects; thus, 3D modeling software has become a necessity for architects. The digital workflow has put computational design—generative design, algorithmic design, and parametric design—into importance. There are emerging trends for the past decade, and a bibliometric analysis can display information about trends in the literature. Literature trends may provide insight into the direction of computational design development. This study uses a bibliometric analysis with VOSviewer and data from Lens to identify the trends from 2011 to 2021. The result indicates several trends: artificial intelligence, computation, machine learning, visualization, and internet technology. The trend analysis needs to be continued in other computational design categories to find continuity in the findings.

show abstract

Generative Design by Reinforcement Learning: Enhancing the Diversity of Topology Optimization Designs

Cited by 47 publications

References 66 publications

Engineering design optimisation using reinforcement learning with episodic controllers

Engineering design optimisation using reinforcement learning with episodic controllers

CreativeGAN: Editing Generative Adversarial Networks for Creative Design Synthesis

Bibliometric Analysis of Generative Design, Algorithmic Design, and Parametric Design in Architecture

Contact Info

Product

Resources

About