KARA: Enhancing High-Dimensional Data Processing with Learnable Activations
Keywords: Kolmogorov-Arnold representation theorem, high-dimensional data processing, learnable activation
TL;DR: This paper presents KARA, an autoencoder leveraging the Kolmogorov-Arnold theorem for improved high-dimensional data compression.
Abstract: In the rapidly advancing field of machine learning, efficiently processing and interpreting high-dimensional data remains a significant challenge. This paper presents the Kolmogorov-Arnold Representation Autoencoder (KARA), a novel autoencoder architecture designed to leverage the Kolmogorov-Arnold representation theorem. By incorporating this mathematical foundation, KARA enhances the representational power and efficiency of neural networks, enabling superior performance in data compression tasks. Experimental results demonstrate that KARA achieves superior performance, positioning it as a promising approach for high-dimensional data processing.
Primary Area: learning theory
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 462
Loading