Searcharxiv⌕ Search

arXiv subjects

Yu Zhao

Publications and source records attributed to Yu Zhao.

At least 199 records · Page 11Linked to original sources
arXiv2023

CPIA Dataset: A Comprehensive Pathological Image Analysis Dataset for Self-supervised Learning Pre-training

Pathological image analysis is a crucial field in computer-aided diagnosis, where deep learning is widely applied. Transfer learning using pre-trained models initialized on natural images has effectively improved the downstream pathological performance. However, the lack of sophisticated domain-specific pathological initialization hinders their potential. Self-supervised learning (SSL) enables pre-training without sample-level labels, which has great potential to overcome the challenge of expensive annotations. Thus, studies focusing on pathological SSL pre-training call for a comprehensive and standardized dataset, similar to the ImageNet in computer vision. This paper presents the comprehensive pathological image analysis (CPIA) dataset, a large-scale SSL pre-training dataset combining 103 open-source datasets with extensive standardization. The CPIA dataset contains 21,427,877 standardized images, covering over 48 organs/tissues and about 100 kinds of diseases, which includes two main data types: whole slide images (WSIs) and characteristic regions of interest (ROIs). A four-scale WSI standardization process is proposed based on the uniform resolution in microns per pixel (MPP), while the ROIs are divided into three scales artificially. This multi-scale dataset is built with the diagnosis habits under the supervision of experienced senior pathologists. The CPIA dataset facilitates a comprehensive pathological understanding and enables pattern discovery explorations. Additionally, to launch the CPIA dataset, several state-of-the-art (SOTA) baselines of SSL pre-training and downstream evaluation are specially conducted. The CPIA dataset along with baselines is available at https://github.com/zhanglab2021/CPIA_Dataset.

eess.IV↗
arXiv2023

Taming Gradient Variance in Federated Learning with Networked Control Variates

Federated learning, a decentralized approach to machine learning, faces significant challenges such as extensive communication overheads, slow convergence, and unstable improvements. These challenges primarily stem from the gradient variance due to heterogeneous client data distributions. To address this, we introduce a novel Networked Control Variates (FedNCV) framework for Federated Learning. We adopt the REINFORCE Leave-One-Out (RLOO) as a fundamental control variate unit in the FedNCV framework, implemented at both client and server levels. At the client level, the RLOO control variate is employed to optimize local gradient updates, mitigating the variance introduced by data samples. Once relayed to the server, the RLOO-based estimator further provides an unbiased and low-variance aggregated gradient, leading to robust global updates. This dual-side application is formalized as a linear combination of composite control variates. We provide a mathematical expression capturing this integration of double control variates within FedNCV and present three theoretical results with corresponding proofs. This unique dual structure equips FedNCV to address data heterogeneity and scalability issues, thus potentially paving the way for large-scale applications. Moreover, we tested FedNCV on six diverse datasets under a Dirichlet distribution with α = 0.1, and benchmarked its performance against six SOTA methods, demonstrating its superiority.

cs.LG↗
arXiv2023

Topological Magnetoresistance of Magnetic Skyrmionic Bubbles

Magnetic skyrmions offer promising prospects for constructing future energy-efficient and high-density information technology, leading to extensive explorations of new skyrmionic materials recently. The topological Hall effect has been widely adopted as a distinctive marker of skyrmion emergence. Alternately, here we propose a novel signature of skyrmion state by quantitatively investigating the magnetoresistance (MR) induced by skyrmionic bubbles in CeMn2Ge2. An intriguing finding was revealed: the anomalous MR measured at different temperatures can be normalized into a single curve, regardless of sample thickness. This behavior can be accurately reproduced by the recent chiral spin textures MR model. Further analysis of the MR anomaly allowed us to quantitatively examine the effective magnetic fields of various scattering channels. Remarkably, the analyses, combined with the Lorentz transmission electronic microscopy results, indicate that the in-plane scattering channel with triplet exchange interactions predominantly governs the magnetotransport in the Bloch-type skyrmionic bubble state. Our results not only provide insights into the quantum correction on MR induced by skyrmionic bubble phase, but also present an electrical probing method for studying chiral spin texture formation, evolution and their topological properties, which opens up exciting possibilities for identifying new skyrmionic materials and advancing the methodology for studying chiral spin textures.

cond-mat.mtrl-sci↗
arXiv2023

Synslator: An Interactive Machine Translation Tool with Online Learning

Interactive machine translation (IMT) has emerged as a progression of the computer-aided translation paradigm, where the machine translation system and the human translator collaborate to produce high-quality translations. This paper introduces Synslator, a user-friendly computer-aided translation (CAT) tool that not only supports IMT, but is adept at online learning with real-time translation memories. To accommodate various deployment environments for CAT services, Synslator integrates two different neural translation models to handle translation memories for online learning. Additionally, the system employs a language model to enhance the fluency of translations in an interactive mode. In evaluation, we have confirmed the effectiveness of online learning through the translation models, and have observed a 13% increase in post-editing efficiency with the interactive functionalities of Synslator. A tutorial video is available at:https://youtu.be/K0vRsb2lTt8.

cs.CL↗
arXiv2023

Raptor Encoding for Low-Latency Concurrent Multi-PDU Session Transmission with Security Consideration in B5G Edge Network

In B5G edge networks, end-to-end low-latency and high-reliability transmissions between edge computing nodes and terminal devices are essential. This paper investigates the queue-aware coding scheduling transmission of randomly arriving data packets, taking into account potential eavesdroppers in edge networks. To address these concerns, we introduce SCLER, a Protocol Data Units (PDU) Raptor-encoded multi-path transmission method that overcomes the challenges of a larger attack surface in Concurrent Multipath Transfer (CMT), excessive delay due to asymmetric delay\&bandwidth, and lack of interaction among PDU session bearers. We propose a secure and reliable transmission scheme based on Raptor encoding and distribution that incorporates a queue length-aware encoding strategy. This strategy is modeled using Constrained Markov Decision Process (CMDP), and we solve the constraint optimization problem of optimal decision-making based on a threshold strategy. Numerical results indicate that SCLER effectively reduces data leakage risks while achieving the optimal balance between delay and reliability, thereby ensuring data security. Importantly, the proposed system is compatible with current mobile networks and demonstrates practical applicability.

cs.NI↗
arXiv2023

Workload Distribution with Rateless Encoding: A Low-Latency Computation Offloading Method within Edge Networks

This paper introduces REDC, a comprehensive strategy for offloading computational tasks within mobile Edge Networks (EN) to Distributed Computing (DC) after Rateless Encoding (RE). Despite the efficiency, reliability, and scalability advantages of distributed computing in ENs, straggler-induced latencies and failures pose significant challenges. Coded distributed computing has gained attention for its efficient redundancy computing, alleviating the impact of stragglers. Yet, current research predominantly focuses on tolerating a predefined number of stragglers with minimal encoding redundancy. Furthermore, nodes within edge networks are characterized by their inherent heterogeneity in computation, communication, and storage capacities, and unpredictable straggler effects and failures. To our knowledge, existing encoding offloading approaches lack a systematic design and unified consideration of these characteristics. REDC addresses these issues by adaptively encoding tasks, then distributing the workload based on node variations. In the face of unpredictability failures, the rateless encoding adaptation provides resilience to dynamic straggler effects. Considering the node heterogeneity and system status, tasks are offloaded to optimal subset "valid" nodes. Load distribution decisions are made based on updates to queuing theory modeling through state feedback. The REDC framework is applicable to EN by improving resource utilization and reducing task sequence execution delays. Experimental results demonstrate our method's effectiveness and resilient performance, maintaining efficacy even in the presence of unstable nodes.

cs.DC↗
arXiv2023

Time-dependent global simulations of a thin accretion disc: the effects of magnetically-driven winds on thermal instability

According to the standard thin disc theory, it is predicted that the radiation-pressure-dominated inner region of a thin disc is thermally unstable, while observations suggest that it is common for a thin disc of more than 0.01 Eddington luminosity to be in a thermally stable state. Previous studies have suggested that magnetically driven winds have the potential to suppress instability. In this work, we implement one-dimensional global simulations of the thin accretion disc to study the effects of magnetically driven winds on thermal instability. The winds play a role in transferring the angular momentum of the disc and cooling the disc. When the mass outflow rate of winds is low, the important role of winds is to transfer the angular momentum and then shorten the outburst period. When the winds have a high mass outflow rate, they can calm down the thermal instability. We also explore the parameter space of the magnetic field strength and the mass loading parameter.

astro-ph.HE↗
arXiv2023

Resolution of the Diagonal on the Root Stacks

In this paper, we give a new constructive proof of the semi-orthogonal decomposition of the derived category of (quasi)-coherent sheaves of root stacks, through an explicit resolution of the diagonal.

math.AG↗
arXiv2023

DNAGPT: A Generalized Pre-trained Tool for Versatile DNA Sequence Analysis Tasks

Pre-trained large language models demonstrate potential in extracting information from DNA sequences, yet adapting to a variety of tasks and data modalities remains a challenge. To address this, we propose DNAGPT, a generalized DNA pre-training model trained on over 200 billion base pairs from all mammals. By enhancing the classic GPT model with a binary classification task (DNA sequence order), a numerical regression task (guanine-cytosine content prediction), and a comprehensive token language, DNAGPT can handle versatile DNA analysis tasks while processing both sequence and numerical data. Our evaluation of genomic signal and region recognition, mRNA abundance regression, and artificial genomes generation tasks demonstrates DNAGPT's superior performance compared to existing models designed for specific downstream tasks, benefiting from pre-training using the newly designed model structure.

q-bio.GN↗
arXiv2023

Revisiting Disentanglement and Fusion on Modality and Context in Conversational Multimodal Emotion Recognition

It has been a hot research topic to enable machines to understand human emotions in multimodal contexts under dialogue scenarios, which is tasked with multimodal emotion analysis in conversation (MM-ERC). MM-ERC has received consistent attention in recent years, where a diverse range of methods has been proposed for securing better task performance. Most existing works treat MM-ERC as a standard multimodal classification problem and perform multimodal feature disentanglement and fusion for maximizing feature utility. Yet after revisiting the characteristic of MM-ERC, we argue that both the feature multimodality and conversational contextualization should be properly modeled simultaneously during the feature disentanglement and fusion steps. In this work, we target further pushing the task performance by taking full consideration of the above insights. On the one hand, during feature disentanglement, based on the contrastive learning technique, we devise a Dual-level Disentanglement Mechanism (DDM) to decouple the features into both the modality space and utterance space. On the other hand, during the feature fusion stage, we propose a Contribution-aware Fusion Mechanism (CFM) and a Context Refusion Mechanism (CRM) for multimodal and context integration, respectively. They together schedule the proper integrations of multimodal and context features. Specifically, CFM explicitly manages the multimodal feature contributions dynamically, while CRM flexibly coordinates the introduction of dialogue contexts. On two public MM-ERC datasets, our system achieves new state-of-the-art performance consistently. Further analyses demonstrate that all our proposed mechanisms greatly facilitate the MM-ERC task by making full use of the multimodal and context features adaptively. Note that our proposed methods have the great potential to facilitate a broader range of other conversational multimodal tasks.

cs.CL↗
arXiv2023

Constructing Holistic Spatio-Temporal Scene Graph for Video Semantic Role Labeling

Video Semantic Role Labeling (VidSRL) aims to detect the salient events from given videos, by recognizing the predict-argument event structures and the interrelationships between events. While recent endeavors have put forth methods for VidSRL, they can be mostly subject to two key drawbacks, including the lack of fine-grained spatial scene perception and the insufficiently modeling of video temporality. Towards this end, this work explores a novel holistic spatio-temporal scene graph (namely HostSG) representation based on the existing dynamic scene graph structures, which well model both the fine-grained spatial semantics and temporal dynamics of videos for VidSRL. Built upon the HostSG, we present a nichetargeting VidSRL framework. A scene-event mapping mechanism is first designed to bridge the gap between the underlying scene structure and the high-level event semantic structure, resulting in an overall hierarchical scene-event (termed ICE) graph structure. We further perform iterative structure refinement to optimize the ICE graph, such that the overall structure representation can best coincide with end task demand. Finally, three subtask predictions of VidSRL are jointly decoded, where the end-to-end paradigm effectively avoids error propagation. On the benchmark dataset, our framework boosts significantly over the current best-performing model. Further analyses are shown for a better understanding of the advances of our methods.

cs.CV↗
arXiv2023

A Noisy-Label-Learning Formulation for Immune Repertoire Classification and Disease-Associated Immune Receptor Sequence Identification

Immune repertoire classification, a typical multiple instance learning (MIL) problem, is a frontier research topic in computational biology that makes transformative contributions to new vaccines and immune therapies. However, the traditional instance-space MIL, directly assigning bag-level labels to instances, suffers from the massive amount of noisy labels and extremely low witness rate. In this work, we propose a noisy-label-learning formulation to solve the immune repertoire classification task. To remedy the inaccurate supervision of repertoire-level labels for a sequence-level classifier, we design a robust training strategy: The initial labels are smoothed to be asymmetric and are progressively corrected using the model's predictions throughout the training process. Furthermore, two models with the same architecture but different parameter initialization are co-trained simultaneously to remedy the known "confirmation bias" problem in the self-training-like schema. As a result, we obtain accurate sequence-level classification and, subsequently, repertoire-level classification. Experiments on the Cytomegalovirus (CMV) and Cancer datasets demonstrate our method's effectiveness and superior performance on sequence-level and repertoire-level tasks.

cs.LG↗
arXiv2023

CellMix: A General Instance Relationship based Method for Data Augmentation Towards Pathology Image Classification

In pathology image analysis, obtaining and maintaining high-quality annotated samples is an extremely labor-intensive task. To overcome this challenge, mixing-based methods have emerged as effective alternatives to traditional preprocessing data augmentation techniques. Nonetheless, these methods fail to fully consider the unique features of pathology images, such as local specificity, global distribution, and inner/outer-sample instance relationships. To better comprehend these characteristics and create valuable pseudo samples, we propose the CellMix framework, which employs a novel distribution-oriented in-place shuffle approach. By dividing images into patches based on the granularity of pathology instances and shuffling them within the same batch, the absolute relationships between instances can be effectively preserved when generating new samples. Moreover, we develop a curriculum learning-inspired, loss-driven strategy to handle perturbations and distribution-related noise during training, enabling the model to adaptively fit the augmented data. Our experiments in pathology image classification tasks demonstrate state-of-the-art (SOTA) performance on 7 distinct datasets. This innovative instance relationship-centered method has the potential to inform general data augmentation approaches for pathology image classification. The associated codes are available at https://github.com/sagizty/CellMix.

cs.CV↗
arXiv2023

Symmetry Fractionalized (Irrationalized) Fusion Rules and Two Domain-Wall Verlinde Formulae

We investigate the composite systems consisting of topological orders separated by gapped domain walls. We derive a pair of domain-wall Verlinde formulae, that elucidate the connection between the braiding of interdomain excitations labeled by pairs of anyons in different domains and quasiparticles in the gapped domain wall with their respective fusion rules. Through explicit non-Abelian examples, we showcase the calculation of such braiding and fusion, revealing that the fusion rules for interdomain excitations are generally fractional or irrational. By investigating the correspondence between composite systems and anyon condensation, we unveil the reason for designating these fusion rules as symmetry fractionalized (irrationalized) fusion rules. Our findings hold promise for applications across various fields, such as topological quantum computation, topological field theory, and conformal field theory.

cond-mat.str-el↗
arXiv2023

Characteristic Properties of a Composite System of Topological Phases Separated by Gapped Domain Walls via an Exactly Solvable Hamiltonian Model

In this paper, we construct an exactly solvable lattice Hamiltonian model to investigate the properties of a composite system consisting of multiple topological orders separated by gapped domain walls. There are interdomain elementary excitations labeled by a pair of anyons in different domains of this system; This system also has elementary excitations with quasiparticles in the gapped domain wall. Each set of elementary excitations corresponds to a basis of the ground states of this composite system on the torus, reflecting that the ground-state degeneracy matches the number of either set of elementary excitations. The characteristic properties of this composite system lie in the basis transformations, represented by the $S$ and $T$ matrices: The $S$ matrix encodes the mutual statistics between interdomain excitations and domain-wall quasiparticles, and the $T$ matrix encapsulates the topological spins of interdomain excitations. Our model realizes a spatial counterpart of a temporal phase transition triggered by anyon condensation, bringing the abstract theory of anyon condensation into manifestable spatial interdomain excitation states.

cond-mat.str-el↗
arXiv2023

Mining Interest Trends and Adaptively Assigning SampleWeight for Session-based Recommendation

Session-based Recommendation (SR) aims to predict users' next click based on their behavior within a short period, which is crucial for online platforms. However, most existing SR methods somewhat ignore the fact that user preference is not necessarily strongly related to the order of interactions. Moreover, they ignore the differences in importance between different samples, which limits the model-fitting performance. To tackle these issues, we put forward the method, Mining Interest Trends and Adaptively Assigning Sample Weight, abbreviated as MTAW. Specifically, we model users' instant interest based on their present behavior and all their previous behaviors. Meanwhile, we discriminatively integrate instant interests to capture the changing trend of user interest to make more personalized recommendations. Furthermore, we devise a novel loss function that dynamically weights the samples according to their prediction difficulty in the current epoch. Extensive experimental results on two benchmark datasets demonstrate the effectiveness and superiority of our method.

cs.IR↗
arXiv2023

Desingulization of Quasi-smooth Derived Schemes and Virtual Fundamental Classes

We prove a desingularization theorem for the quasi-smooth derived scheme, in the sense of Hekking. We also propose the conjecture that the K-theoretic integration of the virtual fundamental class of a quasi-smooth derived scheme could be explained as the K-theoretic integration of wedge powers of non-negative rank perfect complexes with Tor-amplitude [0,1] over smooth schemes.

math.AG↗
arXiv2023

A Generalized vanishing theorem for Blow-ups of Quasi-smooth Stacks

We prove a generalized vanishing theorem for certain quasi-coherent sheaves along the derived blow-ups of quasi-smooth derived Artin stacks. We give four applications of the generalized vanishing theorem: we prove a $K$-theoretic version of the generalized vanishing theorem which verified a conjecture of the author and give a new proof of the $K$-theoretic virtual localization theorem for quasi-smooth derived schemes through the intrinsic blow-up theory of Kiem-Li-Savvas; we prove a desingularization theorem for quasi-smooth derived schemes and give an approximation formula for the virtual fundamental classes; we give a resolution of the diagonal along the projection map of blow-ups of smooth varieties, which strengthens the semi-orthogonal decomposition theorem of Orlov; we illustrate the relation between the generalized vanishing theorem and weak categorifications of quantum loop and toroidal algebra actions on the derived category of Nakajima quiver varieties. We also propose several conjectures related to birational geometry and the $L_{\infty}$-algebroid of Calaque-Căldăraru-Tu.

math.AG↗