Lei Ding
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Lei Ding has a total of 68 co-authors. Here are the top co-authors by number of shared publications:
- Linglong Kong
University of Alberta
2 shared publications - Alla Konnikov
Concordia University of Edmonton, Canada
1 shared publication Bei Jiang
Department of Mathematical and Statistical Sciences University of Alberta {yangdi xchang4 yliu16
1 shared publication- Bei Jiang
Department of Mathematical and Statistical Sciences University of Alberta Edmonton Canada
1 shared publication - Bohuai Xiao
SimpleWay.AI, Shenzhen, China
1 shared publication - Boyu Han
Stanford University, Palo Alto, CA, USA
1 shared publication
Latest Publications
- Open Access
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 7757-7768
VeritasFi : An Adaptable, Multi-tiered RAG Framework for Multi-modal Financial Question Answering
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 7757-7768
Open AccessVeritasFi : An Adaptable, Multi-tiered RAG Framework for Multi-modal Financial Question Answering
- 1 January 2026
- Published by Instats
Generative AI for Statistical Science (Free Seminar)
Coordinated by the Biostatistics section of Statistical Society of Canada, this workshop introduces modern generative AI methods and their use in statistics and biostatistics. The focus is on understanding how these models work, where they are useful in practice, and how to apply them responsibly to real data problems. This workshop was coordinated by the Biostatistics section of Statistical Society of Canada.
- 1 January 2026
- Published by Instats
Generative AI for Statistical Science (Free Seminar)
Coordinated by the Biostatistics section of Statistical Society of Canada, this workshop introduces modern generative AI methods and their use in statistics and biostatistics. The focus is on understanding how these models work, where they are useful in practice, and how to apply them responsibly to real data problems. This workshop was coordinated by the Biostatistics section of Statistical Society of Canada.
- Open Access
- 20 November 2024
- Published by Oxford University Press (OUP) in PNAS Nexus
- Vol. 3 (12) , pgae526
Language in job advertisements and the reproduction of labor force gender and racial segregation
Job advertisements (ads) represent the first point of contact between employers and job seekers. By signaling characteristics expected of an ideal candidate, job ads “gatekeep” the labor force and configure its composition. Meanwhile, labor force composition can also shape the wording of job ads. This study develops a multidimensional inventory of gender and EDI (equality, diversity, inclusion) language in job ads. Applying this inventory, it adopts an instrumental-variable approach to disentangle the reciprocal relationships between gender/EDI language in job ads and labor force gender/racial composition. Drawing on the analysis of 28.6 million job ads in the United Kingdom in combination with labor force statistics between 2018 and 2023, the findings reveal three distinct mechanisms through which the bidirectional interplay between language in job ads and labor force composition (re)produces or disrupts labor force gender/racial segregation. They highlight both the benefits and limitations of intervening in the language used in job ads to help reduce labor force gender/racial segregation.
- 20 November 2024
- Published by Oxford University Press (OUP) in PNAS Nexus
- Vol. 3 (12) , pgae526
Open AccessLanguage in job advertisements and the reproduction of labor force gender and racial segregation
Job advertisements (ads) represent the first point of contact between employers and job seekers. By signaling characteristics expected of an ideal candidate, job ads “gatekeep” the labor force and configure its composition. Meanwhile, labor force composition can also shape the wording of job ads. This study develops a multidimensional inventory of gender and EDI (equality, diversity, inclusion) language in job ads. Applying this inventory, it adopts an instrumental-variable approach to disentangle the reciprocal relationships between gender/EDI language in job ads and labor force gender/racial composition. Drawing on the analysis of 28.6 million job ads in the United Kingdom in combination with labor force statistics between 2018 and 2023, the findings reveal three distinct mechanisms through which the bidirectional interplay between language in job ads and labor force composition (re)produces or disrupts labor force gender/racial segregation. They highlight both the benefits and limitations of intervening in the language used in job ads to help reduce labor force gender/racial segregation.
- 4 September 2024
- Published by Springer Nature in Archaeological and Anthropological Sciences
- Vol. 16 (10) , 1-11
Ageing wolves through crown height measurements and its implications for ageing canids
- 4 September 2024
- Published by Springer Nature in Archaeological and Anthropological Sciences
- Vol. 16 (10) , 1-11
Ageing wolves through crown height measurements and its implications for ageing canids
- Open Access
Gaussian Differential Privacy on Riemannian Manifolds
We develop an advanced approach for extending Gaussian Differential Privacy (GDP) to general Riemannian manifolds. The concept of GDP stands out as a prominent privacy definition that strongly warrants extension to manifold settings, due to its central limit properties. By harnessing the power of the renowned Bishop-Gromov theorem in geometric analysis, we propose a Riemannian Gaussian distribution that integrates the Riemannian distance, allowing us to achieve GDP in Riemannian manifolds with bounded Ricci curvature. To the best of our knowledge, this work marks the first instance of extending the GDP framework to accommodate general Riemannian manifolds, encompassing curved spaces, and circumventing the reliance on tangent space summaries. We provide a simple algorithm to evaluate the privacy budget $μ$ on any one-dimensional manifold and introduce a versatile Markov Chain Monte Carlo (MCMC)-based algorithm to calculate $μ$ on any Riemannian manifold with constant curvature. Through simulations on one of the most prevalent manifolds in statistics, the unit sphere $S^d$, we demonstrate the superior utility of our Riemannian Gaussian mechanism in comparison to the previously proposed Riemannian Laplace mechanism for implementing GDP.Open AccessGaussian Differential Privacy on Riemannian Manifolds
We develop an advanced approach for extending Gaussian Differential Privacy (GDP) to general Riemannian manifolds. The concept of GDP stands out as a prominent privacy definition that strongly warrants extension to manifold settings, due to its central limit properties. By harnessing the power of the renowned Bishop-Gromov theorem in geometric analysis, we propose a Riemannian Gaussian distribution that integrates the Riemannian distance, allowing us to achieve GDP in Riemannian manifolds with bounded Ricci curvature. To the best of our knowledge, this work marks the first instance of extending the GDP framework to accommodate general Riemannian manifolds, encompassing curved spaces, and circumventing the reliance on tangent space summaries. We provide a simple algorithm to evaluate the privacy budget $μ$ on any one-dimensional manifold and introduce a versatile Markov Chain Monte Carlo (MCMC)-based algorithm to calculate $μ$ on any Riemannian manifold with constant curvature. Through simulations on one of the most prevalent manifolds in statistics, the unit sphere $S^d$, we demonstrate the superior utility of our Riemannian Gaussian mechanism in comparison to the previously proposed Riemannian Laplace mechanism for implementing GDP. - 12 July 2023
- Published by Elsevier in Leukemia Research
- Vol. 132, 107354
Changing frontline AML treatment patterns from 2013 to 2022
- 12 July 2023
- Published by Elsevier in Leukemia Research
- Vol. 132, 107354
Changing frontline AML treatment patterns from 2013 to 2022
- Open Access
- 2 February 2023
- Published by Elsevier in Ecotoxicology and Environmental Safety
- Vol. 252, 114603
Antibiotic resistome in groundwater and its association with mountain springs and river
- 2 February 2023
- Published by Elsevier in Ecotoxicology and Environmental Safety
- Vol. 252, 114603
Open AccessAntibiotic resistome in groundwater and its association with mountain springs and river
- Open Access
Reducing Selection Bias in Counterfactual Reasoning for Individual Treatment Effects Estimation
Counterfactual reasoning is an important paradigm applicable in many fields, such as healthcare, economics, and education. In this work, we propose a novel method to address the issue of \textit{selection bias}. We learn two groups of latent random variables, where one group corresponds to variables that only cause selection bias, and the other group is relevant for outcome prediction. They are learned by an auto-encoder where an additional regularized loss based on Pearson Correlation Coefficient (PCC) encourages the de-correlation between the two groups of random variables. This allows for explicitly alleviating selection bias by only keeping the latent variables that are relevant for estimating individual treatment effects. Experimental results on a synthetic toy dataset and a benchmark dataset show that our algorithm is able to achieve state-of-the-art performance and improve the result of its counterpart that does not explicitly model the selection bias.
Open AccessReducing Selection Bias in Counterfactual Reasoning for Individual Treatment Effects Estimation
Counterfactual reasoning is an important paradigm applicable in many fields, such as healthcare, economics, and education. In this work, we propose a novel method to address the issue of \textit{selection bias}. We learn two groups of latent random variables, where one group corresponds to variables that only cause selection bias, and the other group is relevant for outcome prediction. They are learned by an auto-encoder where an additional regularized loss based on Pearson Correlation Coefficient (PCC) encourages the de-correlation between the two groups of random variables. This allows for explicitly alleviating selection bias by only keeping the latent variables that are relevant for estimating individual treatment effects. Experimental results on a synthetic toy dataset and a benchmark dataset show that our algorithm is able to achieve state-of-the-art performance and improve the result of its counterpart that does not explicitly model the selection bias.