Biography
I am a postdoctoral researcher at the Massachusetts Institute of Technology, affiliated with MIT Energy Initiative and Laboratory for Information and Decision Systems. Before joining MIT, I pursued a PhD degree in electrical engineering at the Technical University of Denmark in 2017-2020. In 2019, I also visited the School of Industrial and Systems Engineering of Georgia Tech. In March 2021, I defended my thesis titled Stochastic and Private Energy System Optimization (video, slides). Through my PhD studies, I was most interested in decision-making models under uncertainty, in economics of uncertainty, and most recently in algorithmic privacy, all with applications to energy systems. Today, I broaden this portfolio towards certifying decisions under uncertainty (e.g., a priori performance guarantees for multi-stage stochastic programs), providing privacy guarantees for arbitrary convex optimization programs (moving from linear to quadratic and semidefinite private programming), and developing optimization algorithms to assist machine learning tasks. In 2021, I co-organized enOPTIMAL seminar series to promote research exchange during the pandemic.
- Stochastic optimization
- Electric Power Systems
- Algorithmic privacy
- Energy economics
-
MSc+PhD in Electrical Engineering, 2021
Technical University of Denmark
-
MSc in Economics, 2014
Higher School of Economics (Russia)
-
BSc in Electrical Engineering, 2012
Moscow Power Engineering Institute (Russia)
News
March 2023 Our new differentially private algorithms produce high-quality optimization & ML datasets for power systems research, enabling secure data sharing in energy (
preprint
, GitHub repository).
Update (June 2023): accepted at the IEEE Control Systems Letters with the option to present at the 62nd IEEE Conference on Decision and Control in Singapore!
Dec 2022 Completed MIT Kaufman Teaching Certificate Program
Sep 2022 New preprint on aiding gas network optimization with neural networks is now out!
Update (March 2023): selected for a spotlight talk at the Tackling Climate Change with Machine Learning Workshop at the 2023 International Conference on Learning Representations.
Sep 2022 Our new preprint Privacy-Preserving Convex Optimization: When Differential Privacy Meets Stochastic Programming is out! (paper, tutorials, talk for RSRG group (Caltech))
Jun 2022 Presented our work Multi-Stage Linear Decision Rules for Stochastic Control of Natural Gas Networks with Linepack at the XXII Power System Computational Conference in Porto ( slides )
Jun 2022 I am glad to share our new preprint Multi-Stage Investment Decision Rules for Power Systems with Performance Guarantees (paper, data and code,
slides
)
Update (Feb 2023): accepted for publication in the IEEE Transactions on Power Systems journal!
May 2022 Presented Privacy-Preserving Perturbation of Convex Optimization Programs at the casual seminar series Stats&LIDS Tea Talks at MIT ( slides )
Apr 2022 I won a postdoctoral fellowship co-funded by Marie Skłodowska-Curie Actions and Iberdrola Group to research optimization, learning and privacy for future energy systems at MIT Energy initiative.
Mar 2022 I recieved 2021 Best Paper Award for Differentially Private Optimal Power Flow for Distribution Grids and 2021 Outstanding Reviewer Award from IEEE Transactions on Power Systems .
Mar 2022 We are launching the Spring 2022 edition of the enOPTIMAL: Energy, Optimization and Learning online seminar series (see speakers and seminar logistics here).
Feb 2022 Our paper Multi-Stage Linear Decision Rules for Stochastic Control of Natural Gas Networks with Linepack is accepted at the XXII Power System Computational Conference in Porto! (paper, code)
Oct 2021 Happy to announce the Fall 2021 enOPTIMAL Seminar Series! Check the new lineup here.
Jun 2021 Our paper Stochastic Control and Pricing for Natural Gas Networks is accepted for publication in the IEEE Transactions on Control of Network Systems (CONES). (paper, Anubhav’s talk).
Jun 2021 Presented our paper Differentially Private Optimal Power Flow for Distribution Grids at the PowerTech Conference in Madrid (paper, video).
Jun 2021 Presented our work Linear Decision Rules with Performance Guarantees for Scalable Multi-Stage Generation Planning at the Federal Energy Regulatory Commission workshop ( slides , event).
Apr 2021 We are launching a new online seminar series enOPTIMAL: Energy, Optimization and Learning. Many thanks to all speakers who agreed to kick off this seminar series.
Mar 2021 Defended my Ph.D. thesis titled “Stochastic and Private Energy System Optimization” on Zoom. Thank you all who tuned in and supported me today! (thesis, video, slides).
Publications
Submitted
Dvorkin V., Fioretto N., Van Hentenryck P., Kazempour J. and Pinson P.
Privacy-preserving convex optimization: When differential privacy meets stochastic programming
[preprint, code, presentation]
Journal publications
Dvorkin V. and Botterud A.
Differentially private algorithms for synthetic power system datasets
IEEE Control Systems Letters [
preprint
, code]
Dvorkin V., Mallapragada D. and Botterud A. (2023)
Multi-stage decision rules for power generation & storage investments with performance guarantees
IEEE Transactions on Power Systems [preprint, code, LDR tutorial]
Dvorkin V., Mallapragada D., Botterud A., Kazempour J. and Pinson P. (2022)
Multi-stage linear decision rules for stochastic control of natural gas networks with linepack
Electric Power Systems Research (XXII PSCC edition) [paper, code]
Dvorkin V., Ratha A., Pinson P. and Kazempour J. (2021)
Stochastic control and pricing for natural gas networks
IEEE Transactions on Control of Network Systems [paper, code, presentation]
Dvorkin V., Fioretto F., Van Hentenryck P., Pinson P. and Kazempour J. (2021)
Differentially private optimal power flow for distribution grids
IEEE Transactions on Power Systems [paper, code, presentation]
Dvorkin V., Kazempour J. and Pinson P. (2019)
Electricity market equilibrium under information asymmetry
Operations Research Letters [paper, code]
Dvorkin V., Delikaraoglou S. and Morales J.M. (2019)
Setting reserve requirements to approximate the efficiency of the stochastic dispatch
IEEE Transactions on Power Systems [paper, code]
Conference publications
Zhao D., Dvorkin V., Delikaraoglou S., Lamadrid A. J., Botterud A. (2023)
A scalable bilevel framework for renewable energy scheduling
ACM International Conference on Future Energy Systems (e-Energy
2023) [preprint]
Dvorkin V., Chevalier S. and Chatzivasileiadis S. (2023)
Emission-constrained optimization of gas systems with input-convex neural networks
Tackling Climate Change with Machine Learning Workshop at ICLR 2023 [preprint]
Dvorkin V., Kazempour J. and Pinson P. (2020)
Chance-constrained equilibrium in electricity markets with asymmetric forecasts
International Conference on Probabilistic Methods Applied to Power Systems
[paper]
Dvorkin V., Van Hentenryck P., Kazempour J. and Pinson P. (2020)
Differentially private distributed optimal power flow
IEEE Conference on Decision and Control [paper, code, presentation]
Radoszynski A.M., Dvorkin V. and Pinson P. (2019)
Accommodating bounded rationality in pricing demand response
IEEE Milan PowerTech [paper]
Dvorkin V., Kazempour J., Baringo L. and Pinson P. (2018)
A consensus-ADMM approach for strategic generation investment in electricity markets
IEEE Conference on Decision and Control [paper, code]
Thesis
Stochastic and private energy system optimization
Ph.D. Thesis (2021). Technical University of Denmark [thesis]
Multi-stage strategic investment in CCGTs and wind power units via progressive hedging
M.Sc. Thesis (2017). Technical University of Denmark [thesis]