Searcharxiv⌕ Search

arXiv subjects

Yicheng Xu

Publications and source records attributed to Yicheng Xu.

30 records · Page 2Linked to original sources
arXiv2025

Semantic Shift Estimation via Dual-Projection and Classifier Reconstruction for Exemplar-Free Class-Incremental Learning

Exemplar-Free Class-Incremental Learning (EFCIL) aims to sequentially learn from distinct categories without retaining exemplars but easily suffers from catastrophic forgetting of learned knowledge. While existing EFCIL methods leverage knowledge distillation to alleviate forgetting, they still face two critical challenges: semantic shift and decision bias. Specifically, the embeddings of old tasks shift in the embedding space after learning new tasks, and the classifier becomes biased towards new tasks due to training solely with new data, hindering the balance between old and new knowledge. To address these issues, we propose the Dual-Projection Shift Estimation and Classifier Reconstruction (DPCR) approach for EFCIL. DPCR effectively estimates semantic shift through a dual-projection, which combines a learnable transformation with a row-space projection to capture both task-wise and category-wise shifts. Furthermore, to mitigate decision bias, DPCR employs ridge regression to reformulate a classifier reconstruction process. This reconstruction exploits previous in covariance and prototype of each class after calibration with estimated shift, thereby reducing decision bias. Extensive experiments demonstrate that, on various datasets, DPCR effectively balances old and new tasks, outperforming state-of-the-art EFCIL methods. Our codes are available at https://github.com/RHe502/ICML25-DPCR.

cs.CV↗
arXiv2024

Advancing Cross-domain Discriminability in Continual Learning of Vision-Language Models

Continual learning (CL) with Vision-Language Models (VLMs) has overcome the constraints of traditional CL, which only focuses on previously encountered classes. During the CL of VLMs, we need not only to prevent the catastrophic forgetting on incrementally learned knowledge but also to preserve the zero-shot ability of VLMs. However, existing methods require additional reference datasets to maintain such zero-shot ability and rely on domain-identity hints to classify images across different domains. In this study, we propose Regression-based Analytic Incremental Learning (RAIL), which utilizes a recursive ridge regression-based adapter to learn from a sequence of domains in a non-forgetting manner and decouple the cross-domain correlations by projecting features to a higher-dimensional space. Cooperating with a training-free fusion module, RAIL absolutely preserves the VLM's zero-shot ability on unseen domains without any reference data. Additionally, we introduce Cross-domain Task-Agnostic Incremental Learning (X-TAIL) setting. In this setting, a CL learner is required to incrementally learn from multiple domains and classify test images from both seen and unseen domains without any domain-identity hint. We theoretically prove RAIL's absolute memorization on incrementally learned domains. Experiment results affirm RAIL's state-of-the-art performance in both X-TAIL and existing Multi-domain Task-Incremental Learning settings. The code is released at https://github.com/linghan1997/Regression-based-Analytic-Incremental-Learning.

cs.CV↗
arXiv2024

Preamble Design for Joint Frame Synchronization, Frequency Offset Estimation, and Channel Estimation in Upstream Burst-mode Detection of Coherent PONs

Coherent optics has demonstrated significant potential as a viable solution for achieving 100 Gb/s and higher speeds in single-wavelength passive optical networks (PON). However, upstream burst-mode coherent detection is a major challenge when adopting coherent optics in access networks. To accelerate digital signal processing (DSP) convergence with a minimal preamble length, we propose a novel burst-mode preamble design based on a constant amplitude zero auto-correlation sequence. This design facilitates comprehensive estimation of linear channel effects in the frequency domain, including polarization state rotation, differential group delay, chromatic dispersion, and polarization-dependent loss, providing overall system response information for channel equalization pre-convergence. Additionally, this preamble utilizes the same training unit to jointly achieve three key DSP functions: frame synchronization, frequency offset estimation, and channel estimation. This integration contributes to a significant reduction in the preamble length. The feasibility of the proposed preamble with a length of 272 symbols and corresponding DSP was experimentally verified in a 15 Gbaud coherent system using dual-polarization 16 quadrature amplitude modulation. The experimental results based on this scheme showed a superior performance of the convergence acceleration.

eess.SP↗
arXiv2024

Weighted Envy-Freeness in House Allocation

The classic house allocation problem involves assigning $m$ houses to $n$ agents based on their utility functions, ensuring each agent receives exactly one house. A key criterion in these problems is satisfying fairness constraints such as envy-freeness. We extend this problem by considering agents with arbitrary weights, focusing on the concept of weighted envy-freeness, which has been extensively studied in fair division. We present a polynomial-time algorithm to determine whether weighted envy-free allocations exist and, if so, to compute one. Since weighted envy-free allocations do not always exist, we also investigate the potential of achieving such allocations through the use of subsidies. We provide several characterizations for weighted envy-freeable allocations (allocations that can be turned weighted envy-free by introducing subsidies) and show that they do not always exist, which is different from the unweighted setting. Furthermore, we explore the existence of weighted envy-freeable allocations in specific scenarios and outline the conditions under which they exist.

cs.GT↗
arXiv2023

TACR: A Table-alignment-based Cell-selection and Reasoning Model for Hybrid Question-Answering

Hybrid Question-Answering (HQA), which targets reasoning over tables and passages linked from table cells, has witnessed significant research in recent years. A common challenge in HQA and other passage-table QA datasets is that it is generally unrealistic to iterate over all table rows, columns, and linked passages to retrieve evidence. Such a challenge made it difficult for previous studies to show their reasoning ability in retrieving answers. To bridge this gap, we propose a novel Table-alignment-based Cell-selection and Reasoning model (TACR) for hybrid text and table QA, evaluated on the HybridQA and WikiTableQuestions datasets. In evidence retrieval, we design a table-question-alignment enhanced cell-selection method to retrieve fine-grained evidence. In answer reasoning, we incorporate a QA module that treats the row containing selected cells as context. Experimental results over the HybridQA and WikiTableQuestions (WTQ) datasets show that TACR achieves state-of-the-art results on cell selection and outperforms fine-grained evidence retrieval baselines on HybridQA, while achieving competitive performance on WTQ. We also conducted a detailed analysis to demonstrate that being able to align questions to tables in the cell-selection stage can result in important gains from experiments of over 90\% table row and column selection accuracy, meanwhile also improving output explainability.

cs.CL↗
arXiv2022

Approximate the individually fair k-center with outliers

In this paper, we propose and investigate the individually fair $k$-center with outliers (IF$k$CO). In the IF$k$CO, we are given an $n$-sized vertex set in a metric space, as well as integers $k$ and $q$. At most $k$ vertices can be selected as the centers and at most $q$ vertices can be selected as the outliers. The centers are selected to serve all the not-an-outlier (i.e., served) vertices. The so-called individual fairness constraint restricts that every served vertex must have a selected center not too far way. More precisely, it is supposed that there exists at least one center among its $\lceil (n-q) / k \rceil$ closest neighbors for every served vertex. Because every center serves $(n-q) / k$ vertices on the average. The objective is to select centers and outliers, assign every served vertex to some center, so as to minimize the maximum fairness ratio over all served vertices, where the fairness ratio of a vertex is defined as the ratio between its distance with the assigned center and its distance with a $\lceil (n - q )/k \rceil_{\rm th}$ closest neighbor. As our main contribution, a 4-approximation algorithm is presented, based on which we develop an improved algorithm from a practical perspective.

cs.DS↗
arXiv2020

Too Much Information Kills Information: A Clustering Perspective

Clustering is one of the most fundamental tools in the artificial intelligence area, particularly in the pattern recognition and learning theory. In this paper, we propose a simple, but novel approach for variance-based k-clustering tasks, included in which is the widely known k-means clustering. The proposed approach picks a sampling subset from the given dataset and makes decisions based on the data information in the subset only. With certain assumptions, the resulting clustering is provably good to estimate the optimum of the variance-based objective with high probability. Extensive experiments on synthetic datasets and real-world datasets show that to obtain competitive results compared with k-means method (Llyod 1982) and k-means++ method (Arthur and Vassilvitskii 2007), we only need 7% information of the dataset. If we have up to 15% information of the dataset, then our algorithm outperforms both the k-means method and k-means++ method in at least 80% of the clustering tasks, in terms of the quality of clustering. Also, an extended algorithm based on the same idea guarantees a balanced k-clustering result.

cs.LG↗
arXiv2020

mmWave/THz Channel Estimation Using Frequency-Selective Atomic Norm Minimization

We propose a MIMO channel estimation method for millimeter-wave (mmWave) and terahertz (THz) systems based on frequency-selective atomic norm minimization (FS-ANM). For the strong line-of-sight property of the channel in such high-frequency bands, prior knowledge on the ranges of angles of departure/arrival (AoD/AoA) can be obtained as the prior knowledge, which can be exploited by the proposed channel estimator to improve the estimation accuracy. Simulation results show that the proposed method can achieve considerable performance gain when compared with the existing approaches without incorporating the the strong line-of-sight property.

eess.SP↗
arXiv2020

Online Joint Placement and Allocation of Virtual Network Functions with Heterogeneous Servers

Network Function Virtualization (NFV) is a promising virtualization technology that has the potential to significantly reduce the expenses and improve the service agility. NFV makes it possible for Internet Service Providers (ISPs) to employ various Virtual Network Functions (VNFs) without installing new equipments. One of the most attractive approaches in NFV technology is a so-called Joint Placement and Allocation of Virtual Network Functions (JPA-VNF) which considers the balance between VNF investment with Quality of Services (QoS). We introduce a novel capability function to measure the potential of locating VNF instances for each server in the proposed OJPA-HS model. This model allows the servers in the network to be heterogeneous, at the same time combines and generalizes many classical JPA-VNF models. Despite its NP-hardness, we present a provable best-possible deterministic online algorithm based on dynamic programming (DP). To conquer the high complexity of DP, we propose two additional randomized heuristics, the Las Vegas (LV) and Monte Carlo (MC) randomized algorithms, which performs even as good as DP with much smaller complexity. Besides, MC is a promising heuristic in practice as it has the advantage to deal with big data environment. Extensive numerical experiments are constructed for the proposed algorithms in the paper.

cs.NI↗
arXiv2019

A constant FPT approximation algorithm for hard-capacitated k-means

Hard-capacitated $k$-means (HCKM) is one of the fundamental problems remaining open in combinatorial optimization and data mining areas. In this problem, one is required to partition a given $n$-point set into $k$ disjoint clusters with known capacity so as to minimize the sum of within-cluster variances. It is known to be at least APX-hard and for which most of the work is from a meta heuristic perspective. To the best our knowledge, no constant approximation algorithm or existence proof of such an algorithm is known. As our main contribution, we propose an FPT($k$) algorithm with performance guarantee of $69+ε$ for any HCKM instances in this paper.

cs.DS↗
arXiv2019

Joint Timing Offset and Channel Estimation for Multi-user UFMC Uplink

Universal filtered multi-carrier (UFMC), which groups and filters subcarriers before transmission, is a potential multi-carrier modulation technique investigated for the emerging Machine-Type Communications (MTC). Considering the relaxed timing synchronization requirement of UFMC, we design a novel joint timing synchronization and channel estimation method for multi-user UFMC uplink transmission. Aiming at reducing overhead for higher system performance, the joint estimation problem is formulated using atomic norm minimization that enhances the sparsity of timing offset in the continuous frequency domain. Simulation results show that the proposed method can achieve considerable performance gain, as compared with its counterparts.

eess.SP↗