About

Alexandra Chronopoulou is a Clinical Associate Professor in the Department of Statistics at the Illinois College of Liberal Arts & Sciences. She holds a Ph.D. in Statistics from Purdue University, obtained in 2009, and a Diploma in Applied Mathematics from the National Technical University of Athens, earned in 2004. Her research interests encompass fundamental research in statistics, data science and big data analytics, and quantitative methods in the social sciences. Her specific areas of focus include financial engineering, stochastic modeling and simulation, stochastic systems with long memory, statistical inference for stochastic processes, and sequential experimental design. She has been recognized for her teaching excellence, being ranked as excellent by her students for multiple courses, and has received awards such as the Sharp Outstanding Teaching Award in Industrial Engineering and the Engineering Council Award for Excellence in Advising. Her professional profile indicates a strong commitment to research and education in statistical methods and their applications.

Research topics

• Gastroenterology
• Surgery
• Medicine

Selected publications

• The Influence of Past Experience with Wristbands and Use Exposure on Perceived Usability of Wearable Wrist Worn Sensors by Nurses

  Springer series in design and innovation · 2025-01-01

  book-chapter
  PublisherDOI

• A New Proxy for Estimating the Roughness of Volatility

  Journal of risk and financial management · 2024-03-22 · 3 citations

  articleOpen accessSenior authorCorresponding

  In this paper, we propose a new proxy for the unobserved volatility process that will allow us to better understand and hence model a rough or persistent volatility. Starting with a stochastic volatility model with minimal assumptions on the volatility process, we calibrate the model to options’ data and their sensitivities to obtain an implied volatility process. Starting with this new proxy, we then study the roughness/persistence of the volatility using S&P 500 European put option daily data. We then estimate the Hurst index, i.e., roughness/smoothness parameter, of the volatility with various techniques to find that the volatility does exhibit a rough behavior, even in a low-frequency framework.

  PublisherOA PDFDOI

• What Matters for Model Merging at Scale?

  arXiv (Cornell University) · 2024-10-04

  preprintOpen access

  Model merging aims to combine multiple expert models into a more capable single model, offering benefits such as reduced storage and serving costs, improved generalization, and support for decentralized model development. Despite its promise, previous studies have primarily focused on merging a few small models. This leaves many unanswered questions about the effect of scaling model size and how it interplays with other key factors -- like the base model quality and number of expert models -- , to affect the merged model's performance. This work systematically evaluates the utility of model merging at scale, examining the impact of these different factors. We experiment with merging fully fine-tuned models using 4 popular merging methods -- Averaging, Task~Arithmetic, Dare, and TIES -- across model sizes ranging from 1B-64B parameters and merging up to 8 different expert models. We evaluate the merged models on both held-in tasks, i.e., the expert's training tasks, and zero-shot generalization to unseen held-out tasks. Our experiments provide several new insights about model merging at scale and the interplay between different factors. First, we find that merging is more effective when experts are created from strong base models, i.e., models with good zero-shot performance. Second, larger models facilitate easier merging. Third merging consistently improves generalization capabilities. Notably, when merging 8 large expert models, the merged models often generalize better compared to the multitask trained models. Fourth, we can better merge more expert models when working with larger models. Fifth, different merging methods behave very similarly at larger scales. Overall, our findings shed light on some interesting properties of model merging while also highlighting some limitations. We hope that this study will serve as a reference point on large-scale merging for upcoming research.

  PublisherOA PDFDOI

• Analyzing the Work System Elements Impacting Burnout of Health Care Professionals in a COVID-19 Testing Laboratory

  Springer series in design and innovation · 2023-11-04

  book-chapter
  PublisherDOI

• Delta-hedging in fractional volatility models

  Annals of Finance · 2022-11-09 · 1 citations

  articleSenior authorCorresponding
  PublisherDOI

• Detecting Burnout of Health Care Professionals in a COVID-19 Testing Laboratory

  Proceedings of the Human Factors and Ergonomics Society Annual Meeting · 2022-09-01 · 3 citations

  article

  Health care professionals (HCPs) are frequently exposed to Human Factors/Ergonomics (HFE) issues that result in stress, adversely affecting their health and negatively impacting the quality of care. Chronic stress can result in burnout, with negative implications for individuals, health care organizations, and patients. Current approaches to monitor burnout are reactive and require additional work (e.g., survey completion). In this study, we pilot a methodology using unobtrusive sensors and advanced statistics to bridge this important gap. We collected two types of physiological data - heart rate variability (HRV) and electrodermal activity (EDA) - and measures of perceived workload and burnout from three HCPs in a COVID-19 Testing Laboratory. We identified meaningful relationships between physiological data, workload, and burnout, demonstrating that burnout can be identified proactively using real-time sensor data. Future work will expand the timeframe of data collection and include a larger sample with different types of HCPs.

  PublisherDOI

• Inadequate Rectal Pressure and Insufficient Relaxation and Abdominopelvic Coordination in Defecatory Disorders

  Gastroenterology · 2021 · 26 citations

  • Medicine
  • Gastroenterology
  • Surgery
  PublisherOA PDFDOI

• Improving the Lexical Ability of Pretrained Language Models for Unsupervised Neural Machine Translation

  2021-01-01 · 2 citations

  preprintOpen access1st authorCorresponding

  Alexandra Chronopoulou, Dario Stojanovski, Alexander Fraser. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

  PublisherOA PDFDOI

• The LMU Munich System for the WMT 2020 Unsupervised Machine Translation\n Shared Task

  arXiv (Cornell University) · 2020-10-25 · 1 citations

  preprintOpen access1st authorCorresponding

  This paper describes the submission of LMU Munich to the WMT 2020\nunsupervised shared task, in two language directions, German<->Upper Sorbian.\nOur core unsupervised neural machine translation (UNMT) system follows the\nstrategy of Chronopoulou et al. (2020), using a monolingual pretrained language\ngeneration model (on German) and fine-tuning it on both German and Upper\nSorbian, before initializing a UNMT model, which is trained with online\nbacktranslation. Pseudo-parallel data obtained from an unsupervised statistical\nmachine translation (USMT) system is used to fine-tune the UNMT model. We also\napply BPE-Dropout to the low resource (Upper Sorbian) data to obtain a more\nrobust system. We additionally experiment with residual adapters and find them\nuseful in the Upper Sorbian->German direction. We explore sampling during\nbacktranslation and curriculum learning to use SMT translations in a more\nprincipled way. Finally, we ensemble our best-performing systems and reach a\nBLEU score of 32.4 on German->Upper Sorbian and 35.2 on Upper Sorbian->German.\n

  PublisherOA PDFDOI

• The LMU Munich System for the WMT 2020 Unsupervised Machine Translation Shared Task

  2020-01-01 · 2 citations

  preprintOpen access1st authorCorresponding

  This paper describes the submission of LMU Munich to the WMT 2020 unsupervised shared task, in two language directions, GermanUpper Sorbian. Our core unsupervised neural machine translation (UNMT) system follows the strategy of Chronopoulou et al. (2020), using a monolingual pretrained language generation model (on German) and fine-tuning it on both German and Upper Sorbian, before initializing a UNMT model, which is trained with online backtranslation. Pseudo-parallel data obtained from an unsupervised statistical machine translation (USMT) system is used to fine-tune the UNMT model. We also apply BPE-Dropout to the low resource (Upper Sorbian) data to obtain a more robust system. We additionally experiment with residual adapters and find them useful in the Upper Sorbian->German direction. We explore sampling during backtranslation and curriculum learning to use SMT translations in a more principled way. Finally, we ensemble our best-performing systems and reach a BLEU score of 32.4 on German->Upper Sorbian and 35.2 on Upper Sorbian->German.

  PublisherOA PDFDOI

Recent grants

• Estimation, Sampling, and Simulation for Long-Range Dependent Processes

  NSF · $100k · 2018–2022

Frequent coauthors

• Frédéri Viens

  Rice University

  19 shared

• Ciprian A. Tudor

  17 shared

• Konstantinos Spiliopoulos

  9 shared

• Christos Baziotis

  8 shared

• Alexandros Potamianos

  7 shared

• Dario Stojanovski

  Ludwig-Maximilians-Universität München

  6 shared

• Alexander Fraser

  6 shared

• Georgios Fellouris

  University of Illinois Urbana-Champaign

  3 shared

Awards & honors

• Teacher Ranked as Excellent by their students for IE 400 (Fa…
• Teacher Ranked as Excellent by their students for IE 525 (Sp…
• Teacher Ranked as Excellent by their students for IE 598AC (…
• The Sharp Outstanding Teaching Award in Industrial Engineeri…
• Engineering Council Award for Excellence in Advising (2018-2…