arXiv subjects

Sanghoon Lee

Publications and source records attributed to Sanghoon Lee.

At least 55 records · Page 3Linked to original sources

arXiv2024

Quantum Phase Transitions and Dynamics in Perturbed Flatbands

In recent years, there has been a growing interest in flatband systems which exhibit macroscopic degeneracies. These systems offer a valuable mathematical framework for the extreme sensitivity to perturbations and interactions. This sensitivity unveils a wide variety of exotic and unconventional physical phenomena. Moreover, the progress in their experimental realization contributes to the expanding landscape of exploration in this field. This thesis aims to summarize all the works throughout the Ph.D. program. Firstly, an in-depth exploration was conducted on the impact of weak quasiperiodic perturbations on one-dimensional two-band all-bands-flat lattices. These tight-binding Hamiltonian are diagonalized through a sequence of local unitary transformations. By adjusting the quasiperiodic potential parameters, the key achievement involves finding a critical-to-insulator transition and identifying fractality edges in the flatband systems with quasiperiodic perturbations. Next, the investigation delved into the effects of on-site interactions among hard-core bosons in one- and two-dimensional cross-stitch lattices. One key finding is that groundstate energy primarily depends on compact localized states. Moreover, the presence of barriers of compact localized states trap bosons, leading to the emergence of non-ergodic excitation and Hilbert space fragmentation. Lastly, a compact localized eigenstate of the one-dimensional diamond chain using an electric circuit was successfully generated via local (linear and non-linear) driving. This achievement opens up a versatile circuit platform for the generation of flatbands and holds promise for potential applications in the field of quantum information. I hope these collective efforts have expanded the frontiers of the field and made a meaningful contribution to the scientific community.

cond-mat.dis-nn↗

arXiv2024

Liouville-Type Theorems on the Hyperbolic Space

In this paper, we establish Liouville-type theorems for a one-parameter family of elliptic PDEs on the standard upper half-plane model of the hyperbolic space, under specific geometric assumptions. Our results indicate that the Euclidean half-plane is the only compactification of the hyperbolic space when the scalar curvature of the compactified metric has a designated sign.

math.DG↗

arXiv2023

Camera-Driven Representation Learning for Unsupervised Domain Adaptive Person Re-identification

We present a novel unsupervised domain adaption method for person re-identification (reID) that generalizes a model trained on a labeled source domain to an unlabeled target domain. We introduce a camera-driven curriculum learning (CaCL) framework that leverages camera labels of person images to transfer knowledge from source to target domains progressively. To this end, we divide target domain dataset into multiple subsets based on the camera labels, and initially train our model with a single subset (i.e., images captured by a single camera). We then gradually exploit more subsets for training, according to a curriculum sequence obtained with a camera-driven scheduling rule. The scheduler considers maximum mean discrepancies (MMD) between each subset and the source domain dataset, such that the subset closer to the source domain is exploited earlier within the curriculum. For each curriculum sequence, we generate pseudo labels of person images in a target domain to train a reID model in a supervised way. We have observed that the pseudo labels are highly biased toward cameras, suggesting that person images obtained from the same camera are likely to have the same pseudo labels, even for different IDs. To address the camera bias problem, we also introduce a camera-diversity (CD) loss encouraging person images of the same pseudo label, but captured across various cameras, to involve more for discriminative feature learning, providing person representations robust to inter-camera variations. Experimental results on standard benchmarks, including real-to-real and synthetic-to-real scenarios, demonstrate the effectiveness of our framework.

cs.CV↗

arXiv2023

Improved energy decay estimate for Dir-stationary $Q$-valued functions and its applications

In this paper, we establish an improved decay estimate for the Dirichlet energy of Dir-stationary $Q$-valued functions. As a direct application of this estimate, we derive a Liouville-type theorem for bounded Dir-stationary $Q$-valued functions defined on $\mathbb{R}^m$. Additionally, in an attempt to establish the continuity of Dir-stationary $Q$-valued functions, we confirm that such functions exhibit the Lebesgue property at every point within their domain. Finally, we observe that the improved decay estimate implicates that Dir-stationary $Q$-valued functions reside within a generalized Campanato-Morrey space.

math.DG↗

arXiv2023

Some Uniformization Problems for a Fourth order Conformal Curvature

In this paper, we establish the existence of conformal deformations that uniformize fourth order curvature on 4-dimensional Riemannian manifolds with positive conformal invariants. Specifically, we prove that any closed, compact Riemannian manifold with positive Yamabe invariant and total $Q$-curvature can be conformally deformed into a metric with positive scalar curvature and constant $Q$-curvature. For a Riemannian manifold with umbilic boundary, positive first Yamabe invariant and total $(Q, T)$-curvature, it is possible to deform it into two types of Riemannian manifolds with totally geodesic boundary and positive scalar curvature. The first type satisfies $Q\equiv \text{constant}, T \equiv 0$ while the second type satisfies $Q\equiv 0, T \equiv \text{constant}$.

math.DG↗

arXiv2023

Extracting Structured Seed-Mediated Gold Nanorod Growth Procedures from Literature with GPT-3

Although gold nanorods have been the subject of much research, the pathways for controlling their shape and thereby their optical properties remain largely heuristically understood. Although it is apparent that the simultaneous presence of and interaction between various reagents during synthesis control these properties, computational and experimental approaches for exploring the synthesis space can be either intractable or too time-consuming in practice. This motivates an alternative approach leveraging the wealth of synthesis information already embedded in the body of scientific literature by developing tools to extract relevant structured data in an automated, high-throughput manner. To that end, we present an approach using the powerful GPT-3 language model to extract structured multi-step seed-mediated growth procedures and outcomes for gold nanorods from unstructured scientific text. GPT-3 prompt completions are fine-tuned to predict synthesis templates in the form of JSON documents from unstructured text input with an overall accuracy of $86\%$. The performance is notable, considering the model is performing simultaneous entity recognition and relation extraction. We present a dataset of 11,644 entities extracted from 1,137 papers, resulting in 268 papers with at least one complete seed-mediated gold nanorod growth procedure and outcome for a total of 332 complete procedures.

physics.app-ph↗

arXiv2023

Critical States Generators from Perturbed Flatbands

One-dimensional all-bands-flat lattices are networks with all bands being flat and highly degenerate. They can always be diagonalized by a finite sequence of local unitary transformations parameterized by a set of angles $θ_{i}$. In our previous work, Ref.~\onlinecite{lee2023critical}, we demonstrated that quasiperiodic perturbations of the one-dimensional all-bands-flat lattice with $θ_{i} = π/4$ give rise to a critical-to-insulator transition and fractality edges separating critical from localized states. In this study we consider the full range of angles $θ_{i}$s available for the all-bands-flat model and study the effect of the quasiperiodic perturbation. For weak perturbation, we derive an effective Hamiltonian and we identify the sets of $θ_{i}$s for which the effective model maps to extended or off-diagonal Harper models and hosts critical states. For all the other values of the angles the spectrum is localized. Upon increasing the perturbation strength, the extended Harper model evolves into the system with energy dependent critical-to-insulator transitions, that we dub \emph{fractality edges}. The case where the effective model maps onto the off-diagonal Harper model features a critical-to-insulator transition at a finite disorder strength.

cond-mat.dis-nn↗

arXiv2022

Structured information extraction from complex scientific text with fine-tuned large language models

Intelligently extracting and linking complex scientific information from unstructured text is a challenging endeavor particularly for those inexperienced with natural language processing. Here, we present a simple sequence-to-sequence approach to joint named entity recognition and relation extraction for complex hierarchical information in scientific text. The approach leverages a pre-trained large language model (LLM), GPT-3, that is fine-tuned on approximately 500 pairs of prompts (inputs) and completions (outputs). Information is extracted either from single sentences or across sentences in abstracts/passages, and the output can be returned as simple English sentences or a more structured format, such as a list of JSON objects. We demonstrate that LLMs trained in this way are capable of accurately extracting useful records of complex scientific knowledge for three representative tasks in materials chemistry: linking dopants with their host materials, cataloging metal-organic frameworks, and general chemistry/phase/morphology/application information extraction. This approach represents a simple, accessible, and highly-flexible route to obtaining large databases of structured knowledge extracted from unstructured text. An online demo is available at http://www.matscholar.com/info-extraction.

cs.CL↗

arXiv2022

Decomposed Knowledge Distillation for Class-Incremental Semantic Segmentation

Class-incremental semantic segmentation (CISS) labels each pixel of an image with a corresponding object/stuff class continually. To this end, it is crucial to learn novel classes incrementally without forgetting previously learned knowledge. Current CISS methods typically use a knowledge distillation (KD) technique for preserving classifier logits, or freeze a feature extractor, to avoid the forgetting problem. The strong constraints, however, prevent learning discriminative features for novel classes. We introduce a CISS framework that alleviates the forgetting problem and facilitates learning novel classes effectively. We have found that a logit can be decomposed into two terms. They quantify how likely an input belongs to a particular class or not, providing a clue for a reasoning process of a model. The KD technique, in this context, preserves the sum of two terms (i.e., a class logit), suggesting that each could be changed and thus the KD does not imitate the reasoning process. To impose constraints on each term explicitly, we propose a new decomposed knowledge distillation (DKD) technique, improving the rigidity of a model and addressing the forgetting problem more effectively. We also introduce a novel initialization method to train new classifiers for novel classes. In CISS, the number of negative training samples for novel classes is not sufficient to discriminate old classes. To mitigate this, we propose to transfer knowledge of negatives to the classifiers successively using an auxiliary classifier, boosting the performance significantly. Experimental results on standard CISS benchmarks demonstrate the effectiveness of our framework.

cs.CV↗

arXiv2022

Critical-to-Insulator Transitions and Fractality Edges in Perturbed Flatbands

We study the effect of quasiperiodic perturbations on one-dimensional all-bands-flat lattice models. Such networks can be diagonalized by a finite sequence of local unitary transformations parameterized by angles $θ_i$. Without loss of generality, we focus on the case of two bands with bandgap $Δ$. Weak perturbations lead to an effective Hamiltonian with both on- and off-diagonal quasiperiodic terms that depend on $θ_i$. For some angle values, the effective model coincides with the extended Harper model. By varying the parameters of the quasiperiodic potentials, \iffalse and the manifold angles $θ_i$ \fi we observe localized insulating states and an entire parameter range hosting critical states with subdiffusive transport. For finite quasiperiodic potential strength, the critical-to-insulating transition becomes energy dependent with what we term fractality edges separating localized from critical states.

cond-mat.dis-nn↗

arXiv2022

OIMNet++: Prototypical Normalization and Localization-aware Learning for Person Search

We address the task of person search, that is, localizing and re-identifying query persons from a set of raw scene images. Recent approaches are typically built upon OIMNet, a pioneer work on person search, that learns joint person representations for performing both detection and person re-identification (reID) tasks. To obtain the representations, they extract features from pedestrian proposals, and then project them on a unit hypersphere with L2 normalization. These methods also incorporate all positive proposals, that sufficiently overlap with the ground truth, equally to learn person representations for reID. We have found that 1) the L2 normalization without considering feature distributions degenerates the discriminative power of person representations, and 2) positive proposals often also depict background clutter and person overlaps, which could encode noisy features to person representations. In this paper, we introduce OIMNet++ that addresses the aforementioned limitations. To this end, we introduce a novel normalization layer, dubbed ProtoNorm, that calibrates features from pedestrian proposals, while considering a long-tail distribution of person IDs, enabling L2 normalized person representations to be discriminative. We also propose a localization-aware feature learning scheme that encourages better-aligned proposals to contribute more in learning discriminative representations. Experimental results and analysis on standard person search benchmarks demonstrate the effectiveness of OIMNet++.

cs.CV↗

arXiv2022

Text-mined dataset of gold nanoparticle synthesis procedures, morphologies, and size entities

Gold nanoparticles are highly desired for a range of technological applications due to their tunable properties, which are dictated by the size and shape of the constituent particles. Many heuristic methods for controlling the morphological characteristics of gold nanoparticles are well known. However, the underlying mechanisms controlling their size and shape remain poorly understood, partly due to the immense range of possible combinations of synthesis parameters. Data-driven methods can offer insight to help guide understanding of these underlying mechanisms, so long as sufficient synthesis data are available. To facilitate data mining in this direction, we have constructed and made publicly available a dataset of codified gold nanoparticle synthesis protocols and outcomes extracted directly from the nanoparticle materials science literature using natural language processing and text-mining techniques. This dataset contains 5,154 data records, each representing a single gold nanoparticle synthesis article, filtered from a database of 4,973,165 publications. Each record contains codified synthesis protocols and extracted morphological information from a total of 7,608 experimental and 12,519 characterization paragraphs.

cond-mat.mtrl-sci↗

arXiv2021

Learning by Aligning: Visible-Infrared Person Re-identification using Cross-Modal Correspondences

We address the problem of visible-infrared person re-identification (VI-reID), that is, retrieving a set of person images, captured by visible or infrared cameras, in a cross-modal setting. Two main challenges in VI-reID are intra-class variations across person images, and cross-modal discrepancies between visible and infrared images. Assuming that the person images are roughly aligned, previous approaches attempt to learn coarse image- or rigid part-level person representations that are discriminative and generalizable across different modalities. However, the person images, typically cropped by off-the-shelf object detectors, are not necessarily well-aligned, which distract discriminative person representation learning. In this paper, we introduce a novel feature learning framework that addresses these problems in a unified way. To this end, we propose to exploit dense correspondences between cross-modal person images. This allows to address the cross-modal discrepancies in a pixel-level, suppressing modality-related features from person representations more effectively. This also encourages pixel-wise associations between cross-modal local features, further facilitating discriminative feature learning for VI-reID. Extensive experiments and analyses on standard VI-reID benchmarks demonstrate the effectiveness of our approach, which significantly outperforms the state of the art.

cs.CV↗

arXiv2021

HVPR: Hybrid Voxel-Point Representation for Single-stage 3D Object Detection

We address the problem of 3D object detection, that is, estimating 3D object bounding boxes from point clouds. 3D object detection methods exploit either voxel-based or point-based features to represent 3D objects in a scene. Voxel-based features are efficient to extract, while they fail to preserve fine-grained 3D structures of objects. Point-based features, on the other hand, represent the 3D structures more accurately, but extracting these features is computationally expensive. We introduce in this paper a novel single-stage 3D detection method having the merit of both voxel-based and point-based features. To this end, we propose a new convolutional neural network (CNN) architecture, dubbed HVPR, that integrates both features into a single 3D representation effectively and efficiently. Specifically, we augment the point-based features with a memory module to reduce the computational cost. We then aggregate the features in the memory, semantically similar to each voxel-based one, to obtain a hybrid 3D representation in a form of a pseudo image, allowing to localize 3D objects in a single stage efficiently. We also propose an Attentive Multi-scale Feature Module (AMFM) that extracts scale-aware features considering the sparse and irregular patterns of point clouds. Experimental results on the KITTI dataset demonstrate the effectiveness and efficiency of our approach, achieving a better compromise in terms of speed and accuracy.

cs.CV↗

arXiv2020

Unsupervised Learning of Deep-Learned Features from Breast Cancer Images

Detecting cancer manually in whole slide images requires significant time and effort on the laborious process. Recent advances in whole slide image analysis have stimulated the growth and development of machine learning-based approaches that improve the efficiency and effectiveness in the diagnosis of cancer diseases. In this paper, we propose an unsupervised learning approach for detecting cancer in breast invasive carcinoma (BRCA) whole slide images. The proposed method is fully automated and does not require human involvement during the unsupervised learning procedure. We demonstrate the effectiveness of the proposed approach for cancer detection in BRCA and show how the machine can choose the most appropriate clusters during the unsupervised learning procedure. Moreover, we present a prototype application that enables users to select relevant groups mapping all regions related to the groups in whole slide images.

eess.IV↗

arXiv2020

HistomicsML2.0: Fast interactive machine learning for whole slide imaging data

Extracting quantitative phenotypic information from whole-slide images presents significant challenges for investigators who are not experienced in developing image analysis algorithms. We present new software that enables rapid learn-by-example training of machine learning classifiers for detection of histologic patterns in whole-slide imaging datasets. HistomicsML2.0 uses convolutional networks to be readily adaptable to a variety of applications, provides a web-based user interface, and is available as a software container to simplify deployment.

q-bio.QM↗

arXiv2019

GraphX-Convolution for Point Cloud Deformation in 2D-to-3D Conversion

In this paper, we present a novel deep method to reconstruct a point cloud of an object from a single still image. Prior arts in the field struggle to reconstruct an accurate and scalable 3D model due to either the inefficient and expensive 3D representations, the dependency between the output and number of model parameters or the lack of a suitable computing operation. We propose to overcome these by deforming a random point cloud to the object shape through two steps: feature blending and deformation. In the first step, the global and point-specific shape features extracted from a 2D object image are blended with the encoded feature of a randomly generated point cloud, and then this mixture is sent to the deformation step to produce the final representative point set of the object. In the deformation process, we introduce a new layer termed as GraphX that considers the inter-relationship between points like common graph convolutions but operates on unordered sets. Moreover, with a simple trick, the proposed model can generate an arbitrary-sized point cloud, which is the first deep method to do so. Extensive experiments verify that we outperform existing models and halve the state-of-the-art distance score in single image 3D reconstruction.

cs.CV↗

arXiv2019

HATS: A Hierarchical Graph Attention Network for Stock Movement Prediction

Many researchers both in academia and industry have long been interested in the stock market. Numerous approaches were developed to accurately predict future trends in stock prices. Recently, there has been a growing interest in utilizing graph-structured data in computer science research communities. Methods that use relational data for stock market prediction have been recently proposed, but they are still in their infancy. First, the quality of collected information from different types of relations can vary considerably. No existing work has focused on the effect of using different types of relations on stock market prediction or finding an effective way to selectively aggregate information on different relation types. Furthermore, existing works have focused on only individual stock prediction which is similar to the node classification task. To address this, we propose a hierarchical attention network for stock prediction (HATS) which uses relational data for stock market prediction. Our HATS method selectively aggregates information on different relation types and adds the information to the representations of each company. Specifically, node representations are initialized with features extracted from a feature extraction module. HATS is used as a relational modeling module with initialized node representations. Then, node representations with the added information are fed into a task-specific layer. Our method is used for predicting not only individual stock prices but also market index movements, which is similar to the graph classification task. The experimental results show that performance can change depending on the relational data used. HATS which can automatically select information outperformed all the existing methods.

q-fin.ST↗