Modern Evolution Strategies for Creativity: Fitting Concrete Images and Abstract Concepts
By Yingtao Tian, David R Ha. 2021
63 citations recorded by the source corpus.
Evolutionary algorithms have been used in the digital art scene since the 1970s. A popular application of genetic algorithms is to optimize the procedural placement of vector graphic primitives to resemble a given painting. In recent years, deep learning-based approaches have also been proposed to generate procedural drawings, which can be optimized using gradient descent. In this work, we revisit the use of evolutionary algorithms for computational creativity. We find that modern evolution strategies (ES) algorithms, when tasked with the placement of shapes, offer large improvements in both quality and efficiency compared to traditional genetic algorithms, and even comparable to gradient-based methods. We demonstrate that ES is also well suited at optimizing the placement of shapes to fit the CLIP model, and can produce diverse, distinct geometric abstractions that are aligned with human interpretation of language. Videos and demo: https://es-clip.github.io/
Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Geology, Generative Models, Optimization & Training
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computational Mechanics
In Generative Models
In Optimization & Training
Citations in the corpus
- VQGAN-CLIP: Open Domain Image Generation and Editing with Natural Language Guidance (2022, background): “[25], and Tian and Ha [46] create additional components (see Section 2.”
- Solving Inverse Problems in Stochastic Self-Organising Systems through Invariant Representations (2025, unknown): “Instead of encoding only low-level information, such as colour and intensity, embeddings represent higher-level visual attributes, such as shape, texture, structure, and semantic content [27–31].”