1 code implementation • 2 May 2024 • Kevin Ma, Daniele Grandi, Christopher McComb, Kosa Goucher-Lambert
Specifically, LLMs are used to generate a total of 4, 000 design solutions across five distinct design topics, eight combinations of parameters, and eight different types of prompt engineering techniques, comparing each combination of parameter and prompt engineering method across four different diversity metrics.
1 code implementation • 30 May 2023 • Kevin Ma, Daniele Grandi, Christopher McComb, Kosa Goucher-Lambert
Expert evaluations indicate that the LLM-generated solutions have higher average feasibility and usefulness while the crowdsourced solutions have more novelty.
no code implementations • 16 May 2023 • Shahroz Khan, Panagiotis Kaklis, Kosa Goucher-Lambert
However, the SAEM and AEM produce better-performing designs.
no code implementations • 29 Apr 2023 • Shahroz Khan, Kosa Goucher-Lambert, Konstantinos Kostas, Panagiotis Kaklis
In this work, we introduce ShipHullGAN, a generic parametric modeller built using deep convolutional generative adversarial networks (GANs) for the versatile representation and generation of ship hulls.