Researcher profile

Taehun Kim

Taehun Kim contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
9works
0followers
9topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

9 published item(s)

preprint2022arXiv

A Style-aware Discriminator for Controllable Image Translation

Current image-to-image translations do not control the output domain beyond the classes used during training, nor do they interpolate between different domains well, leading to implausible results. This limitation largely arises because labels do not consider the semantic distance. To mitigate such problems, we propose a style-aware discriminator that acts as a critic as well as a style encoder to provide conditions. The style-aware discriminator learns a controllable style space using prototype-based self-supervised learning and simultaneously guides the generator. Experiments on multiple datasets verify that the proposed model outperforms current state-of-the-art image-to-image translation methods. In contrast with current methods, the proposed approach supports various applications, including style interpolation, content transplantation, and local image translation.

preprint2022arXiv

Hubble selection of the weak scale from QCD quantum critical point

There is growing evidence that the small weak scale may be related to self-organized criticality. In this regard, we note that if the strange quark were lighter, the QCD phase transition could have been first order, possibly exhibiting quantum critical points at zero temperature as a function of the Higgs vacuum expectation value $v_h$ smaller than (but near) the weak scale. We show that these quantum critical points allow a dynamical selection of the observed weak scale, via quantum-dominated stochastic evolutions of the value of $v_h$ during eternal inflation. Although the values of $v_h$ in different Hubble patches are described by a probability distribution in the multiverse, inflationary quantum dynamics ensures that the peak of the distribution evolves toward critical points (self-organized criticality), driven mainly by the largest Hubble expansion rate there -- the Hubble selection of the universe. To this end, we first explore the quantum critical points of the three-flavor QCD linear sigma model, parametrized by $v_h$ at zero temperature, and we present a relaxion model for the weak scale. Among the patches that have reached reheating, it results in a sharp probability distribution of $v_h$ near the observed weak scale, which is critical not to the crossover at $v_h=0$ but to the sharp transition at ${\sim}Λ_{\rm QCD}$.

preprint2021arXiv

Fast and Accurate Transferability Measurement for Heterogeneous Multivariate Data

Given a set of heterogeneous source datasets with their classifiers, how can we quickly find the most useful source dataset for a specific target task? We address the problem of measuring transferability between source and target datasets, where the source and the target have different feature spaces and distributions. We propose Transmeter, a fast and accurate method to estimate the transferability of two heterogeneous multivariate datasets. We address three challenges in measuring transferability between two heterogeneous multivariate datasets: reducing time, minimizing domain gap, and extracting meaningful homogeneous representations. To overcome the above issues, we utilize a pre-trained source model, an adversarial network, and an encoder-decoder architecture. Extensive experiments on heterogeneous multivariate datasets show that Transmeter gives the most accurate transferability measurement with up to 10.3 times faster performance than its competitor. We also show that selecting the best source data with Transmeter followed by a full transfer leads to the best transfer accuracy and the fastest running time.

preprint2021arXiv

Pressure-induced transition from Jeff=1/2 to S=1/2 states in CuAl2O4

The spin-orbit entangled (SOE) Jeff-state has been a fertile ground to study novel quantum phenomena. Contrary to the conventional weakly correlated Jeff=1/2 state of 4d and 5d transition metal compounds, the ground state of CuAl2O4 hosts a Jeff=1/2 state with a strong correlation of Coulomb U. Here, we report that surprisingly Cu2+ ions of CuAl2O4 overcome the otherwise usually strong Jahn-Teller distortion and instead stabilize the SOE state, although the cuprate has relatively small spin-orbit coupling. From the x-ray absorption spectroscopy and high-pressure x-ray diffraction studies, we obtained definite evidence of the Jeff=1/2 state with a cubic lattice at ambient pressure. We also found the pressure-induced structural transition to a compressed tetragonal lattice consisting of the spin-only S=1/2 state for pressure higher than Pc=8 GPa. This phase transition from the Mott insulating Jeff=1/2 to the S=1/2 states is a unique phenomenon and has not been reported before. Our study offers a rare example of the SOE Jeff-state under strong electron correlation and its pressure-induced transition to the S=1/2 state.

preprint2020arXiv

Centaur: A Chiplet-based, Hybrid Sparse-Dense Accelerator for Personalized Recommendations

Personalized recommendations are the backbone machine learning (ML) algorithm that powers several important application domains (e.g., ads, e-commerce, etc) serviced from cloud datacenters. Sparse embedding layers are a crucial building block in designing recommendations yet little attention has been paid in properly accelerating this important ML algorithm. This paper first provides a detailed workload characterization on personalized recommendations and identifies two significant performance limiters: memory-intensive embedding layers and compute-intensive multi-layer perceptron (MLP) layers. We then present Centaur, a chiplet-based hybrid sparse-dense accelerator that addresses both the memory throughput challenges of embedding layers and the compute limitations of MLP layers. We implement and demonstrate our proposal on an Intel HARPv2, a package-integrated CPU+FPGA device, which shows a 1.7-17.2x performance speedup and 1.7-19.5x energy-efficiency improvement than conventional approaches.

preprint2020arXiv

Constraining the gravitational coupling of axion dark matter at LIGO

The axion-gravity Chern-Simons coupling is well motivated but is relatively weakly constrained, partly due to difficult measurements of gravity. We study the sensitivity of LIGO measurements of chirping gravitational waves (GWs) on such coupling. When the frequency of the propagating GW matches with that of the coherent oscillation of axion dark matter field, the decay of axions into gravitons can be stimulated, resonantly enhancing the GW. Such a resonance peak can be detected at LIGO as a deviation from the chirping waveform. Since all observed GWs will undergo similar resonant enhancement from the Milky-Way (MW) axion halo, LIGO O1+O2 observations can potentially provide the strongest constraint on the coupling, at least for the axion mass $m_a = 5 \times 10^{-13} - 5 \times 10^{-12}$ eV. Along the course, we also emphasize the relevance of the finite coherence of axion fields and the ansatz separating forward and backward propagations of GWs. As a result, the parity violation of the Chern-Simons coupling is not observable from chirping GWs.

preprint2020arXiv

GRB lensing parallax: Closing primordial black hole dark matter mass window

The primordial black hole (PBH) comprising full dark matter (DM) abundance is currently allowed if its mass lies between $10^{-16}M_{\odot} \lesssim M \lesssim 10^{-11} M_{\odot}$. This lightest mass range is hard to be probed by ongoing gravitational lensing observations. In this paper, we advocate that an old idea of the lensing parallax of Gamma-ray bursts (GRBs), observed simultaneously by spatially separated detectors, can probe the unconstrained mass range; and that of nearby stars can probe a heavier mass range. In addition to various good properties of GRBs, astrophysical separations achievable around us --- $r_\oplus \text{--}$ AU --- is just large enough to resolve the GRB lensing by lightest PBH DM.

preprint2020arXiv

Momentum-dependent magnon lifetime in the metallic non-collinear triangular antiferromagnet CrB2

Non-collinear magnetic order arises for various reasons in several magnetic systems and exhibits interesting spin dynamics. Despite its ubiquitous presence, little is known of how magnons, otherwise stable quasiparticles, decay in these systems, particularly in metallic magnets. Using inelastic neutron scattering, we examine the magnetic excitation spectra in a metallic non-collinear antiferromagnet CrB$_{2}$, in which Cr atoms form a triangular lattice and display incommensurate magnetic order. Our data show intrinsic magnon damping and continuum-like excitations that cannot be explained by linear spin wave theory. The intrinsic magnon linewidth $Γ(q,E_{q})$ shows very unusual momentum dependence, which our analysis shows to originate from the combination of two-magnon decay and the Stoner continuum. By comparing the theoretical predictions with the experiments, we identify where in the momentum and energy space one of the two factors becomes more dominant. Our work constitutes a rare comprehensive study of the spin dynamics in metallic non-collinear antiferromagnets. It reveals, for the first time, definite experimental evidence of the higher-order effects in metallic antiferromagnets.

preprint2020arXiv

Probing Cosmic Strings with Gravitational-Wave Fringe

Cosmic strings are important remnants of early-Universe phase transitions. We show that they can be probed by Gravitational Waves (GWs) from compact binary mergers. If such chirping GW passes by a cosmic string, it is gravitationally lensed and left with a characteristic signal of the lensing -- the GW fringe. It is observable naturally through the frequency chirping of GWs. This allows to probe cosmic strings with small tension $Δ= 8πG μ= 10^{-6} \text{ -- } 10^{-10}$, just below the current constraint, at high-frequency LIGO-band and mid-band detectors. Although its detection rates are estimated to be small, even a single detection can be used to identify a cosmic string. Contrary to the stochastic GW produced from loop decays only in local $U(1)$ models, the GW fringe can directly probe straight strings model independently. This is also complementary to the existing probes with the strong lensing of light.