Showing 1–5 of 5 results for author: Lee, S G

Uniform Transformation: Refining Latent Representation in Variational Autoencoders

Authors: Ye Shi, C. S. George Lee

Abstract: Irregular distribution in latent space causes posterior collapse, misalignment between posterior and prior, and ill-sampling problem in Variational Autoencoders (VAEs). In this paper, we introduce a novel adaptable three-stage Uniform Transformation (UT) module -- Gaussian Kernel Density Estimation (G-KDE) clustering, non-parametric Gaussian Mixture (GM) Modeling, and Probability Integral Transfor… ▽ More Irregular distribution in latent space causes posterior collapse, misalignment between posterior and prior, and ill-sampling problem in Variational Autoencoders (VAEs). In this paper, we introduce a novel adaptable three-stage Uniform Transformation (UT) module -- Gaussian Kernel Density Estimation (G-KDE) clustering, non-parametric Gaussian Mixture (GM) Modeling, and Probability Integral Transform (PIT) -- to address irregular latent distributions. By reconfiguring irregular distributions into a uniform distribution in the latent space, our approach significantly enhances the disentanglement and interpretability of latent representations, overcoming the limitation of traditional VAE models in capturing complex data structures. Empirical evaluations demonstrated the efficacy of our proposed UT module in improving disentanglement metrics across benchmark datasets -- dSprites and MNIST. Our findings suggest a promising direction for advancing representation learning techniques, with implication for future research in extending this framework to more sophisticated datasets and downstream tasks. △ Less

Submitted 2 July, 2024; originally announced July 2024.

Comments: Accepted by 2024 IEEE 20th International Conference on Automation Science and Engineering

The Hyperinvarinat Subspace Problem

Authors: Sa Ge Lee

Abstract: The hyperinvariant subspace problem is solved in the setting of Hilbert and right Hamilton space, motivated by my earlier works in the invariant subspace problem. The hyperinvariant subspace problem is solved in the setting of Hilbert and right Hamilton space, motivated by my earlier works in the invariant subspace problem. △ Less

Submitted 24 June, 2023; originally announced June 2023.

Comments: I thank deeply Prof. Charles Akemann (UCSB), Prof. Sang Hoon Lee (CNU), Heon Lee (SNU), and my wife Soon Hee Kim

MSC Class: 47A15

The Invariant Subspace Problem

Authors: Sa Ge Lee

Abstract: The invariant subspace problem is solved correcting my earlier attempts [6]-[12]. The invariant subspace problem is solved correcting my earlier attempts [6]-[12]. △ Less

Submitted 24 June, 2023; v1 submitted 25 February, 2020; originally announced February 2020.

Comments: I thank deeply Prof. Charles Akemann (UCSB), Prof. Sang Hoon Lee (CNU), Heon Lee (SNU), and my wife Soon Hee Kim

MSC Class: 47A15

The Classification of Factor Group Von Neumann Algebras

Authors: Sa Ge Lee

Abstract: The groups distinguish their von Neumann algebras, in the case when these are factors. The groups distinguish their von Neumann algebras, in the case when these are factors. △ Less

Submitted 20 May, 2015; v1 submitted 13 May, 2015; originally announced May 2015.

Comments: This paper has been withdrawn by the author. The author found the critical fault starting at Lemma0.1 and hence subsequent claims from Lemma0.2 up to the main theorem0.6. So, I want to withdraw my paper "The groups distinguish their von Neumann algebras, in the case when these are factors"

MSC Class: 46L10

Multi-Robot Control Using Time-Varying Density Functions

Authors: Sung G. Lee, Magnus Egerstedt

Abstract: This paper presents an approach to externally influencing a team of robots by means of time-varying density functions. These density functions represent rough references for where the robots should be located. To this end, a continuous-time algorithm is proposed that moves the robots so as to provide optimal coverage given the density functions as they evolve over time. The developed algorithm rep… ▽ More This paper presents an approach to externally influencing a team of robots by means of time-varying density functions. These density functions represent rough references for where the robots should be located. To this end, a continuous-time algorithm is proposed that moves the robots so as to provide optimal coverage given the density functions as they evolve over time. The developed algorithm represents an extension to previous coverage algorithms in that time-varying densities are explicitly taken into account in a provable manner. A distributed approximation to this algorithm is moreover proposed whereby the robots only need to access information from adjacent robots. Simulations and robotic experiments show that the proposed algorithms do indeed exhibit the desired behaviors in practice as well as in theory. △ Less

Submitted 1 April, 2014; originally announced April 2014.