arXiv subjects

Xiaoqing Tan

Publications and source records attributed to Xiaoqing Tan.

16 recordsLinked to original sources

arXiv2025

Double Machine Learning Methods for Estimating Average Treatment Effects: A Comparative Study

Observational cohort studies are increasingly being used for comparative effectiveness research to assess the safety of therapeutics. Recently, various doubly robust methods have been proposed for average treatment effect estimation by combining the treatment model and the outcome model via different vehicles, such as matching, weighting, and regression. The key advantage of doubly robust estimators is that they require either the treatment model or the outcome model to be correctly specified to obtain a consistent estimator of average treatment effects, and therefore lead to a more accurate and often more precise inference. However, little work has been done to understand how doubly robust estimators differ due to their unique strategies of using the treatment and outcome models and how machine learning techniques can be combined to boost their performance, which we call double machine learning estimators. Here we examine multiple popular doubly robust methods and compare their performance using different treatment and outcome modeling via extensive simulations and a real-world application. We found that incorporating machine learning with doubly robust estimators such as the targeted maximum likelihood estimator gives the best overall performance. Practical guidance on how to apply doubly robust estimators is provided.

stat.ME↗

arXiv2023

ROBBIE: Robust Bias Evaluation of Large Generative Language Models

As generative large language models (LLMs) grow more performant and prevalent, we must develop comprehensive enough tools to measure and improve their fairness. Different prompt-based datasets can be used to measure social bias across multiple text domains and demographic axes, meaning that testing LLMs on more datasets can potentially help us characterize their biases more fully, and better ensure equal and equitable treatment of marginalized demographic groups. In this work, our focus is two-fold: (1) Benchmarking: a comparison of 6 different prompt-based bias and toxicity metrics across 12 demographic axes and 5 families of generative LLMs. Out of those 6 metrics, AdvPromptSet and HolisticBiasR are novel datasets proposed in the paper. The comparison of those benchmarks gives us insights about the bias and toxicity of the compared models. Therefore, we explore the frequency of demographic terms in common LLM pre-training corpora and how this may relate to model biases. (2) Mitigation: we conduct a comprehensive study of how well 3 bias/toxicity mitigation techniques perform across our suite of measurements. ROBBIE aims to provide insights for practitioners while deploying a model, emphasizing the need to not only measure potential harms, but also understand how they arise by characterizing the data, mitigate harms once found, and balance any trade-offs. We open-source our analysis code in hopes of encouraging broader measurements of bias in future LLMs.

cs.CL↗

arXiv2023

Quantum multiparameter estimation with graph states

In the SU(2) dynamics, it is especially significant to achieve a simultaneous optimal multiparameter estimation but it is very difficult. Evolution on SU(N) dynamics is a research method to explore simultaneous multiparameter estimation with the quantum network. As the highly entangled states, graph state, is an intrinsical quantum resource for quantum metrology. For n-qubit graph state, we propose a simultaneous multiparameter estimation scheme that investigates evolution in SU(N) dynamics. For single-parameter estimation, the precision limit beyond the Heisenberg limit in the higher dimension spin of SU(2). We consider two scenarios where the Hamiltonian operator is commutation and non-commutation respectively and verify that the global estimation precision is higher than the local estimation precision. In the parameter limit condition, the precision of parameter estimation for the simultaneous estimation of each parameter is equal to the precision of the singleparameter estimation. In addition, we find a precision-enhancement scheme that depends on the dynamics SU(N). The smaller the N for the dynamics evolution, the higher the precision of the parameter estimation. Finally, we prove that the graph state is the optimal state of quantum metrology, a set of optimal measurement basic can be found, and the precision limit of multiparameter estimation can attain the quantum Cramér-Rao bound.

quant-ph↗

arXiv2023

Causal Inference under Data Restrictions

This dissertation focuses on modern causal inference under uncertainty and data restrictions, with applications to neoadjuvant clinical trials, distributed data networks, and robust individualized decision making. In the first project, we propose a method under the principal stratification framework to identify and estimate the average treatment effects on a binary outcome, conditional on the counterfactual status of a post-treatment intermediate response. Under mild assumptions, the treatment effect of interest can be identified. We extend the approach to address censored outcome data. The proposed method is applied to a neoadjuvant clinical trial and its performance is evaluated via simulation studies. In the second project, we propose a tree-based model averaging approach to improve the estimation accuracy of conditional average treatment effects at a target site by leveraging models derived from other potentially heterogeneous sites, without them sharing subject-level data. The performance of this approach is demonstrated by a study of the causal effects of oxygen therapy on hospital survival rates and backed up by comprehensive simulations. In the third project, we propose a robust individualized decision learning framework with sensitive variables to improve the worst-case outcomes of individuals caused by sensitive variables that are unavailable at the time of decision. Unlike most existing work that uses mean-optimal objectives, we propose a robust learning framework by finding a newly defined quantile- or infimum-optimal decision rule. From a causal perspective, we also generalize the classic notion of (average) fairness to conditional fairness for individual subjects. The reliable performance of the proposed method is demonstrated through synthetic experiments and three real-data applications.

stat.ME↗

arXiv2022

RISE: Robust Individualized Decision Learning with Sensitive Variables

This paper introduces RISE, a robust individualized decision learning framework with sensitive variables, where sensitive variables are collectible data and important to the intervention decision, but their inclusion in decision making is prohibited due to reasons such as delayed availability or fairness concerns. A naive baseline is to ignore these sensitive variables in learning decision rules, leading to significant uncertainty and bias. To address this, we propose a decision learning framework to incorporate sensitive variables during offline training but not include them in the input of the learned decision rule during model deployment. Specifically, from a causal perspective, the proposed framework intends to improve the worst-case outcomes of individuals caused by sensitive variables that are unavailable at the time of decision. Unlike most existing literature that uses mean-optimal objectives, we propose a robust learning framework by finding a newly defined quantile- or infimum-optimal decision rule. The reliable performance of the proposed method is demonstrated through synthetic experiments and three real-world applications.

cs.LG↗

arXiv2022

A Tree-based Model Averaging Approach for Personalized Treatment Effect Estimation from Heterogeneous Data Sources

Accurately estimating personalized treatment effects within a study site (e.g., a hospital) has been challenging due to limited sample size. Furthermore, privacy considerations and lack of resources prevent a site from leveraging subject-level data from other sites. We propose a tree-based model averaging approach to improve the estimation accuracy of conditional average treatment effects (CATE) at a target site by leveraging models derived from other potentially heterogeneous sites, without them sharing subject-level data. To our best knowledge, there is no established model averaging approach for distributed data with a focus on improving the estimation of treatment effects. Specifically, under distributed data networks, our framework provides an interpretable tree-based ensemble of CATE estimators that joins models across study sites, while actively modeling the heterogeneity in data sources through site partitioning. The performance of this approach is demonstrated by a real-world study of the causal effects of oxygen therapy on hospital survival rate and backed up by comprehensive simulation results.

stat.ML↗

arXiv2022

Verification of colorable hypergraph states with stabilizer test

Many-body quantum states, as a matter of fact, are extremely essential to solve certain mathematical problems or simulate quantum systems in measurement-based quantum computation. However, how to verify large scale quantum states, such as hypergraph states, is an exceedingly hard task for multi-body quantum systems. Here, we propose a novel fault-tolerant solution for verification of colorable hypergraph states by using stabilizer test. Compared with the adaptive stabilizer test, our protocol is dramatically facilitating by making only Pauli-X and Pauli-Z measurement. As to appliance, it will be also applied to blind quantum computing.

quant-ph↗

arXiv2022

Identifying Principal Stratum Causal Effects Conditional on a Post-treatment Intermediate Response

In neoadjuvant trials on early-stage breast cancer, patients are usually randomized into a control group and a treatment group with an additional target therapy. Early efficacy of the new regimen is assessed via the binary pathological complete response (pCR) and the eventual efficacy is assessed via long-term clinical outcomes such as survival. Although pCR is strongly associated with survival, it has not been confirmed as a surrogate endpoint. To fully understand its clinical implication, it is important to establish causal estimands such as the causal effect in survival for patients who would achieve pCR under the new regimen. Under the principal stratification framework, previous studies focus on sensitivity analyses by varying model parameters in an imposed model on counterfactual outcomes. Under mild assumptions, we propose an approach to estimate those model parameters using empirical data and subsequently the causal estimand of interest. We also extend our approach to address censored outcome data. The proposed method is applied to a recent clinical trial and its performance is evaluated via simulation studies.

stat.ME↗

arXiv2022

Point3D: tracking actions as moving points with 3D CNNs

Spatio-temporal action recognition has been a challenging task that involves detecting where and when actions occur. Current state-of-the-art action detectors are mostly anchor-based, requiring sensitive anchor designs and huge computations due to calculating large numbers of anchor boxes. Motivated by nascent anchor-free approaches, we propose Point3D, a flexible and computationally efficient network with high precision for spatio-temporal action recognition. Our Point3D consists of a Point Head for action localization and a 3D Head for action classification. Firstly, Point Head is used to track center points and knot key points of humans to localize the bounding box of an action. These location features are then piped into a time-wise attention to learn long-range dependencies across frames. The 3D Head is later deployed for the final action classification. Our Point3D achieves state-of-the-art performance on the JHMDB, UCF101-24, and AVA benchmarks in terms of frame-mAP and video-mAP. Comprehensive ablation studies also demonstrate the effectiveness of each module proposed in our Point3D.

cs.CV↗

arXiv2020

Towards Improving Spatiotemporal Action Recognition in Videos

Spatiotemporal action recognition deals with locating and classifying actions in videos. Motivated by the latest state-of-the-art real-time object detector You Only Watch Once (YOWO), we aim to modify its structure to increase action detection precision and reduce computational time. Specifically, we propose four novel approaches in attempts to improve YOWO and address the imbalanced class issue in videos by modifying the loss function. We consider two moderate-sized datasets to apply our modification of YOWO - the popular Joint-annotated Human Motion Data Base (J-HMDB-21) and a private dataset of restaurant video footage provided by a Carnegie Mellon University-based startup, Agot.AI. The latter involves fast-moving actions with small objects as well as unbalanced data classes, making the task of action localization more challenging. We implement our proposed methods in the GitHub repository https://github.com/stoneMo/YOWOv2.

cs.CV↗

arXiv2020

Parameter-controlled inserting constructions of constant dimension subspace codes

A basic problem in constant dimension subspace coding is to determine the maximal possible size ${\bf A}_q(n,d,k)$ of a set of $k$-dimensional subspaces in ${\bf F}_q^n$ such that the subspace distance satisfies $\operatorname{dis}(U,V)=2k-2\dim(U \cap V) \geq d$ for any two different $k$-dimensional subspaces $U$ and $V$ in this set. In this paper we propose new parameter-controlled inserting constructions of constant dimension subspace codes. These inserting constructions are flexible because they are controlled by parameters. Several new better lower bounds which are better than all previously constructive lower bounds can be derived from our flexible inserting constructions. $141$ new constant dimension subspace codes of distances $4,6,8$ better than previously best known codes are constructed.

cs.IT↗

arXiv2019

A Hybrid and Universal Blind Quantum Computation

In blind quantum computation (BQC), a client delegates her quantum computation to a server with universal quantum computers who learns nothing about the client's private information. In measurement-based BQC model, entangled states are generally used to realize quantum computing. However, to generate a large-scale entangled state in experiment becomes a challenge issue. In circuit-based BQC model, single-qubit gates can be realized precisely, but entangled gates are probabilistically successful. This remains a challenge to realize entangled gates with a deterministic method in some systems. To solve above two problems, we propose the first hybrid universal BQC protocol based on measurements and circuits, where the client prepares single-qubit states and the server performs universal quantum computing. We analyze and prove the correctness, blindness and verifiability of the proposed protocol.

quant-ph↗

arXiv2019

Local distinguishability of quantum states in bipartite systems

In this article, we show a sufficient and necessary condition for locally distinguishable bipartite states via one-way local operations and classical communication (LOCC). With this condition, we present some minimal structures of one-way LOCC indistinguishable quantum state sets. As long as an indistinguishable subset exists in a state set, the set is not distinguishable. We also list several distinguishable sets as instances.

quant-ph↗

arXiv2016

Capacity-achieving and Flicker-free FEC coding scheme for Dimmable Visible Light Communication Based on Polar Codes

Visible light communication (VLC) could provide short-range optical wireless communication together with illumination using LED lighting. However, conventional forward error correction (FEC) codes for reliable communication do not have the features for dimming support and flicker mitigation which are required in VLC for the main functionality of lighting. Therefore, auxiliary coding techniques are usually needed, which eventually reduce the coding efficiency and increase the complexity. In this paper, a polar codes-based FEC coding scheme for dimmable VLC is proposed to increase the coding efficiency and simplify the coding structure. Experimental results show that the proposed scheme has the following advantages: 1) equal probability of 1's and 0's in codewords, which is inherently supporting 50% dimming balance; 2) short run length property (about 90% bits have runs shorter than 5) which can avoid flickers and additional run-length limited line coding; 3) higher coding efficiency about twofold than that of other coding schemes; 4) capacity achieving error correction performance with low-complexity encoding and decoding, which is about 3 dB higher coding gain than that of RS(64,32) in IEEE standard for dimming ratio 50% and about 1 dB higher coding gain than that of LDPC codes for dimming ratio 25% (or 75%).

cs.IT↗

arXiv2013

Postprocessing for quantum random number generators: entropy evaluation and randomness extraction

Quantum random-number generators (QRNGs) can offer a means to generate information-theoretically provable random numbers, in principle. In practice, unfortunately, the quantum randomness is inevitably mixed with classical randomness due to classical noises. To distill this quantum randomness, one needs to quantify the randomness of the source and apply a randomness extractor. Here, we propose a generic framework for evaluating quantum randomness of real-life QRNGs by min-entropy, and apply it to two different existing quantum random-number systems in the literature. Moreover, we provide a guideline of QRNG data postprocessing for which we implement two information-theoretically provable randomness extractors: Toeplitz-hashing extractor and Trevisan's extractor.

quant-ph↗

arXiv2012

Explicit combinatorial design

A combinatorial design is a family of sets that are almost disjoint, which is applied in pseudo random number generations and randomness extractions. The parameter, $ρ$, quantifying the overlap between the sets within the family, is directly related to the length of a random seed needed and the efficiency of an extractor. Nisan and Wigderson proposed an explicit construction of designs in 1994. Later in 2003, Hartman and Raz proved a bound of $ρ\le e^2$ for the Nisan-Wigderson construction in a limited parameter regime. In this work, we prove a tighter bound of $ρ<e$ with the entire parameter range by slightly refining the Nisan-Wigderson construction. Following the block idea used by Raz, Reingold, and Vadhan, we present an explicit weak design with $ρ=1$.

math.CO↗