Computer Science > Neural and Evolutionary Computing
[Submitted on 4 Feb 2022 (v1), revised 7 Feb 2022 (this version, v2), latest version 6 Jun 2022 (v4)]
Title:COIL: Constrained Optimization in Learned Latent Space -- Learning Representations for Valid Solutions
View PDFAbstract:Constrained optimization problems can be difficult because their search spaces have properties not conducive to search, e.g., multimodality, discontinuities, or deception. To address such difficulties, considerable research has been performed on creating novel evolutionary algorithms or specialized genetic operators. However, if the representation that defined the search space could be altered such that it only permitted valid solutions that satisfied the constraints, the task of finding the optimal would be made more feasible without any need for specialized optimization algorithms. We propose the use of a Variational Autoencoder to learn such representations. We present Constrained Optimization in Latent Space (COIL), which uses a VAE to generate a learned latent representation from a dataset comprising samples from the valid region of the search space according to a constraint, thus enabling the optimizer to find the objective in the new space defined by the learned representation. We investigate the value of this approach on different constraint types and for different numbers of variables. We show that, compared to an identical GA using a standard representation, COIL with its learned latent representation can satisfy constraints and find solutions with distance to objective up to two orders of magnitude closer.
Submission history
From: Dr Peter J. Bentley [view email][v1] Fri, 4 Feb 2022 14:45:37 UTC (2,355 KB)
[v2] Mon, 7 Feb 2022 11:16:23 UTC (2,356 KB)
[v3] Wed, 13 Apr 2022 15:32:33 UTC (1,688 KB)
[v4] Mon, 6 Jun 2022 12:05:52 UTC (1,493 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.