Abstract

AbstractThis paper proposes a Reinforcement Learning (RL) based design approach that augments existing algorithmic generative processes through the emergence of a form of artificial design intuition. The research presented in the paper is embedded within a highly speculative research project, Artificial Agency, exploring the operation of Machine Learning (ML) in generative design and digital fabrication. After describing the inherent limitations of contemporary generative design processes, the paper compares the three fundamental types of machine learning frameworks in terms of their characteristics and potential impact on generative design. A theoretical framework is defined to demonstrate the methodology of integrating RL with existing generative design procedures, which is further explained with a Random Walk based experimental design example. The paper includes detailed RL definitions as well as critical reflections on its impact and the effects of its implementation. The proposed artificial intuition within this generative approach is currently being further developed through a series of ongoing and proposed research trajectories noted in the conclusion. The ambition of this research is to deepen the integration of intention with machine learning in generative design.

Highlights

  • Architectural design has been fundamentally impacted over the past three decades by the integration of emerging technologies and processual theory which have contributed to the proliferation of generative design methodologies [1]

  • Framed by the limitation of contemporary generative methodologies, this paper proposes a Reinforcement Learning based approach to integrate Machine Learning with current computational design processes

  • The research presented in the paper is part of an ongoing research project, Artificial Agency, with aims to explore the operations of machine learning with generative design and autonomous fabrication process, undertaken at the RMIT Architecture, Snooks Research lab

Read more

Summary

Introduction

Architectural design has been fundamentally impacted over the past three decades by the integration of emerging technologies and processual theory which have contributed to the proliferation of generative design methodologies [1]. Framed by the limitation of contemporary generative methodologies, this paper proposes a Reinforcement Learning based approach to integrate Machine Learning with current computational design processes. This will be demonstrated through a simple design experiment based on a random walk algorithm. The broader ambition of this research is to leverage the generative potential of contemporary processes while integrating intuition through machine learning

Contemporary Algorithmic Generative System
Artificial Intuitions
Machine Learning with Generative Design
Reinforcement Learning Reinforcement
Methodology
Intuitive Random Walk Formation
RL Actions Definition
RL Reward Definition
Training Process and Outcomes The Random
Further Research
Conclusions
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call