arXiv subjects

Lina Lin

Publications and source records attributed to Lina Lin.

4 recordsLinked to original sources

arXiv2023

Reducing Symbiosis Bias Through Better A/B Tests of Recommendation Algorithms

It is increasingly common in digital environments to use A/B tests to compare the performance of recommendation algorithms. However, such experiments often violate the stable unit treatment value assumption (SUTVA), particularly SUTVA's "no hidden treatments" assumption, due to the shared data between algorithms being compared. This results in a novel form of bias, which we term "symbiosis bias," where the performance of each algorithm is influenced by the training data generated by its competitor. In this paper, we investigate three experimental designs--cluster-randomized, data-diverted, and user-corpus co-diverted experiments--aimed at mitigating symbiosis bias. We present a theoretical model of symbiosis bias and simulate the impact of each design in dynamic recommendation environments. Our results show that while each design reduces symbiosis bias to some extent, they also introduce new challenges, such as reduced training data in data-diverted experiments. We further validate the existence of symbiosis bias using data from a large-scale A/B test conducted on a global recommender system, demonstrating that symbiosis bias affects treatment effect estimates in the field. Our findings provide actionable insights for researchers and practitioners seeking to design experiments that accurately capture algorithmic performance without bias in treatment effect estimates introduced by shared data.

stat.ME↗

arXiv2022

Verifying Integrity of Deep Ensemble Models by Lossless Black-box Watermarking with Sensitive Samples

With the widespread use of deep neural networks (DNNs) in many areas, more and more studies focus on protecting DNN models from intellectual property (IP) infringement. Many existing methods apply digital watermarking to protect the DNN models. The majority of them either embed a watermark directly into the internal network structure/parameters or insert a zero-bit watermark by fine-tuning a model to be protected with a set of so-called trigger samples. Though these methods work very well, they were designed for individual DNN models, which cannot be directly applied to deep ensemble models (DEMs) that combine multiple DNN models to make the final decision. It motivates us to propose a novel black-box watermarking method in this paper for DEMs, which can be used for verifying the integrity of DEMs. In the proposed method, a certain number of sensitive samples are carefully selected through mimicking real-world DEM attacks and analyzing the prediction results of the sub-models of the non-attacked DEM and the attacked DEM on the carefully crafted dataset. By analyzing the prediction results of the target DEM on these carefully crafted sensitive samples, we are able to verify the integrity of the target DEM. Different from many previous methods, the proposed method does not modify the original DEM to be protected, which indicates that the proposed method is lossless. Experimental results have shown that the DEM integrity can be reliably verified even if only one sub-model was attacked, which has good potential in practice.

cs.CR↗

arXiv2019

Statistical significance in high-dimensional linear mixed models

This paper concerns the development of an inferential framework for high-dimensional linear mixed effect models. These are suitable models, for instance, when we have $n$ repeated measurements for $M$ subjects. We consider a scenario where the number of fixed effects $p$ is large (and may be larger than $M$), but the number of random effects $q$ is small. Our framework is inspired by a recent line of work that proposes de-biasing penalized estimators to perform inference for high-dimensional linear models with fixed effects only. In particular, we demonstrate how to correct a `naive' ridge estimator in extension of work by B\"uhlmann (2013) to build asymptotically valid confidence intervals for mixed effect models. We validate our theoretical results with numerical experiments, in which we show our method outperforms those that fail to account for correlation induced by the random effects. For a practical demonstration we consider a riboflavin production dataset that exhibits group structure, and show that conclusions drawn using our method are consistent with those obtained on a similar dataset without group structure.

stat.ME↗

arXiv2015

Estimation of High-Dimensional Graphical Models Using Regularized Score Matching

Graphical models are widely used to model stochastic dependences among large collections of variables. We introduce a new method of estimating undirected conditional independence graphs based on the score matching loss, introduced by Hyvarinen (2005), and subsequently extended in Hyvarinen (2007). The regularized score matching method we propose applies to settings with continuous observations and allows for computationally efficient treatment of possibly non-Gaussian exponential family models. In the well-explored Gaussian setting, regularized score matching avoids issues of asymmetry that arise when applying the technique of neighborhood selection, and compared to existing methods that directly yield symmetric estimates, the score matching approach has the advantage that the considered loss is quadratic and gives piecewise linear solution paths under $\ell_1$ regularization. Under suitable irrepresentability conditions, we show that $\ell_1$-regularized score matching is consistent for graph estimation in sparse high-dimensional settings. Through numerical experiments and an application to RNAseq data, we confirm that regularized score matching achieves state-of-the-art performance in the Gaussian case and provides a valuable tool for computationally efficient estimation in non-Gaussian graphical models.

stat.ME↗