Draft:Maria Dimakopoulou

Maria Dimakopoulou
NationalityGreek
Alma materStanford University
Known forReinforcement Learning, Contextual Bandits, Causal Inference
Awards"Arvanitidis" Stanford Graduate Fellowship, Onassis Foundation Graduate Fellowship
Scientific career
FieldsMachine Learning, Reinforcement Learning, Contextual Bandits, Causal Inference
InstitutionsNetflix, Stanford University
Doctoral advisorBenjamin Van Roy, Susan Athey

Maria Dimakopoulou is a Greek computer scientist and Senior Research Scientist at Spotify[1], specializing in machine learning, particularly in the areas of reinforcement learning, contextual bandits, and causal inference.

Early Life and Education

edit

After graduating with a Dipl. Ing. from the School of Electrical and Computer Engineering[2] with a perfect 10/10 GPA[3], National Technical University of Athens, Dimakopoulou completed her PhD in Management Science & Engineering at Stanford University[4]. She was advised by Benjamin Van Roy and Susan Athey. Her PhD research was funded by the "Arvanitidis" Stanford Graduate Fellowship[5] in Memory of William K. Linvill and the Onassis Foundation Graduate Fellowship.

Career and Research

edit

Dimakopoulou's research[6] focuses on the development and application of machine learning algorithms. She has made significant contributions to reinforcement learning and contextual bandits, which are critical for decision-making processes under uncertainty.

Key Contributions

edit
  • Reinforcement Learning and Contextual Bandits: Developed algorithms for contextual bandit problems, which are used in recommendation systems and personalized healthcare treatments.
  • Causal Inference: Worked on methods to ensure robust predictions and decisions from observational data, addressing challenges in adaptive data collection.
  • Innovative Algorithms and Estimators: Co-authored influential papers on off-policy evaluation and risk minimization, improving the reliability of machine learning models.

Selected Publications

edit
  • Bibaut, A., Dimakopoulou, M., Kallus, N., Chambaz, A., & van der Laan, M. (2021). Post-Contextual-Bandit Inference. NeurIPS.
  • Bibaut, A., Kallus, N., Dimakopoulou, M., Chambaz, A., & van der Laan, M. (2021). Risk Minimization from Adaptively Collected Data. NeurIPS.
  • Su, Y., Dimakopoulou, M., Krishnamurthy, A., & Dudik, M. (2020). Doubly Robust Off-Policy Evaluation with Shrinkage. ICML.
  • Dimakopoulou, M., Vlassis, N., & Jebara, T. (2019). Marginal Posterior Sampling for Slate Bandits. IJCAI.

Awards and Recognition

edit
  • "Arvanitidis" Stanford Graduate Fellowship in Memory of William K. Linvill
  • Onassis Foundation Graduate Fellowship


References

edit
  1. ^ Maria Dimakopoulou's Homepage, https://mdimakopoulou.github.io
  2. ^ Dimakopoulou, M., Improving Reliability & Efficiency Of Performance Monitoring In Linux, Diploma Thesis, National Technical University of Athens, http://dx.doi.org/10.26240/heal.ntua.5508
  3. ^ Huffington Post, Techs and the City: Part 2 Femmes Up, https://www.huffpost.com/entry/techs-and-the-city-part-2_b_9589002
  4. ^ Dimakopoulou, M., Coordinated exploration in concurrent reinforcement learning, PhD Thesis, Stanford University, https://purl.stanford.edu/hs944xz0420
  5. ^ Maria Dimakopoulou at Stanford University, https://explorecourses.stanford.edu/instructor/madima)
  6. ^ M. Dimakopoulou Google Scholar profile, https://scholar.google.com/citations?user=ySLrpsYAAAAJ&hl=en