arXiv subjects

Shuaida He

Publications and source records attributed to Shuaida He.

5 recordsLinked to original sources

arXiv2026

Federated LoRA Fine-Tuning for LLMs via Collaborative Alignment

Low-rank adaptation (LoRA) has emerged as a powerful tool for parameter-efficient fine-tuning of large language models (LLMs). This paper studies LoRA under a federated learning setting, enabling collaborative fine-tuning across clients while preserving parameter efficiency. We focus on a highly heterogeneous regime in which clients share only partial structure and a substantial subset may be contaminated. We propose Collaborative Low-rank Alignment and Identifiable Recovery (CLAIR), a contamination-aware framework that relies only on preliminary local estimators. Its formulation applies broadly, from linear regression to neural network and LLM modules, whenever local adaptation can be represented by matrix-valued updates. CLAIR recovers the shared LoRA subspace and detects contaminated clients via a structured low-rank plus block-sparse decomposition. We prove exact recovery of the shared LoRA subspace in the noiseless case, stable recovery under preliminary estimation error, and consistent collaborative-set recovery under mild separation conditions. We further quantify the gain from CLAIR refinement: it reduces off-subspace estimation error through cross-client averaging while preserving client-specific variation within the shared LoRA subspace, thus improves over local fine-tuning whenever this oracle gain outweighs the costs of subspace estimation and benign-client heterogeneity. Empirically, we demonstrate the benefits of CLAIR by fine-tuning a Transformer architecture on a text-copying task. The results show accurate contamination detection and improved benign-client performance compared with local fine-tuning and non-robust federated averaging.

stat.ML↗

arXiv2026

Large-Scale Asset Selection via Metric Dependence with Enriched High Frequency Information

Large-scale portfolio choice is highly sensitive to estimation error, making the preliminary asset selection essential in empirical implementation. Existing selection rules typically rely on scalar returns or low dimensional high frequency summaries, and thus discard intraday risk dynamics that may be relevant for risk adjusted allocation. We propose Metric Dependence Screening (MDS), an asset selection procedure that incorporates high frequency information as object valued data. Each asset day observation is represented as a point-curve object combining daily return with an intraday risk state curve, equipped with a weighted product metric that preserves both reward information and within day risk dynamics. MDS ranks assets by a Fr\'echet variation based dependence score, measuring how much a risk adjusted target explains the metric dispersion of the asset representations. This yields a simple two stage portfolio procedure: MDS first reduces the investable universe, and standard mean-variance or minimum variance allocation is then applied. We develop a target slicing estimator and establish concentration, sure selection, and rank consistency guarantees under $\alpha$-mixing time series dependence and ultrahigh dimensionality. Simulations show that MDS performs well across both Euclidean and non-Euclidean settings. Using high frequency data for $2938$ Chinese A-share stocks from July 2023 to December 2025, we demonstrate that MDS improves out of sample portfolio performance over return based and scalar dependence based benchmarks, highlighting the value of preserving intraday risk dynamics.

stat.AP↗

arXiv2026

The Fr\'echet correlation coefficient for heterogeneous random objects

Modern regression analysis often involves responses and predictors taking values in the same or distinct metric spaces. To rank non-Euclidean heterogeneous predictors in regression by explanatory strength, analogous to the classical $R^2$, we introduce the Fr\'echet correlation coefficient (FCC), defined as the relative reduction in the Fr\'echet variance of the response after conditioning on a specific predictor. FCC is directional, model-free, and interpretable on a unit-scale, attaining one under almost sure functional dependence and zero when the Fr\'echet mean is invariant to conditioning. We propose a novel partition-based estimator that avoids explicit nonparametric estimation of the conditional Fr\'echet mean function, thereby improving both computational efficiency and flexibility. A tailored wild bootstrap algorithm is further developed for testing the Fr\'echet conditional mean dependence. We establish asymptotic theory and evaluate power through extensive simulations.

stat.ME↗

arXiv2024

Large-scale metric objects filtering for binary classification with application to abnormal brain connectivity detection

The classification of random objects within metric spaces without a vector structure has attracted increasing attention. However, the complexity inherent in such non-Euclidean data often restricts existing models to handle only a limited number of features, leaving a gap in real-world applications. To address this, we propose a data-adaptive filtering procedure to identify informative features from large-scale random objects, leveraging a novel Kolmogorov-Smirnov-type statistic defined on the metric space. Our method, applicable to data in general metric spaces with binary labels, exhibits remarkable flexibility. It enjoys a model-free property, as its implementation does not rely on any specified classifier. Theoretically, it controls the false discovery rate while guaranteeing the sure screening property. Empirically, equipped with a Wasserstein metric, it demonstrates superior sample performance compared to Euclidean competitors. When applied to analyze a dataset on autism, our method identifies significant brain regions associated with the condition. Moreover, it reveals distinct interaction patterns among these regions between individuals with and without autism, achieved by filtering hundreds of thousands of covariance matrices representing various brain connectivities.

stat.ME↗

arXiv2023

Differentially private sliced inverse regression in the federated paradigm

Sliced inverse regression (SIR), which includes linear discriminant analysis (LDA) as a special case, is a popular and powerful dimension reduction tool. In this article, we extend SIR to address the challenges of decentralized data, prioritizing privacy and communication efficiency. Our approach, named as federated sliced inverse regression (FSIR), facilitates collaborative estimation of the sufficient dimension reduction subspace among multiple clients, solely sharing local estimates to protect sensitive datasets from exposure. To guard against potential adversary attacks, FSIR further employs diverse perturbation strategies, including a novel vectorized Gaussian mechanism that guarantees differential privacy at a low cost of statistical accuracy. Additionally, FSIR naturally incorporates a collaborative variable screening step, enabling effective handling of high-dimensional client data. Theoretical properties of FSIR are established for both low-dimensional and high-dimensional settings, supported by extensive numerical experiments and real data analysis.

stat.ME↗