EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
2734. Wasserstein Distance Based Market for Differentially-Private Data Trading
Invited abstract in session WA-35: Data Valuation from Data-driven Optimization, stream Stochastic, Robust and Distributionally Robust Optimization.
Wednesday, 8:30-10:00Room: 44 (building: 303A)
Authors (first author is the speaker)
| 1. | Fei Teng
|
| Electrical and Electronic Engineering, Imperial College London | |
| 2. | Saurab Chhachhi
|
| Imperial College London |
Abstract
Data is increasingly vital to deal with the increased intermittency and limited controlability of renewable and distributed energy resources. Concurrently, data sharing raises privacy concerns motivating the need for privacy-enhancing techniques such as differential privacy. Data markets provide a means to enable wider access as well as determine the appropriate privacy-utility trade-off. Existing data market frameworks either require a trusted entity to perform computationally expensive valuations or are unable to capture the combinatorial nature of data value and do not endogenously model the effect of differential privacy. We address these shortcomings by proposing a valuation mechanism based on the Wasserstein Distance for differentially-private data, and two procurement mechanisms leveraging incentive mechanism design theory for task-agnostic data procurement, and task-specific joint task and data optimisation. The latter are reformulated into tractable mixed-integer second-order cone programs. The framework is applied to develop a joint energy and data market. We consider the retailer energy procurement problem where consumers’ demand is uncertain and historical demand data is differentially-private. This is modelled as an integrated forecasting and optimisation problem providing a means of valuing data directly rather than forecasts or accuracy. The value of joint energy/data clearing are highlighted through extensive numerical studies using real smart meter data.
Keywords
- Electricity Markets
- Forecasting
- Engineering Optimization
Status: accepted
Back to the list of papers