Communications of the ACM
sing Randomness to Protect Election Integrity
Credit: iStockPhoto.com
Vanderbilt University professor Eugene Vorobeychik and colleagues have combined game theory and computer security expertise to design a method for auditing electoral outcomes to maximize the odds of revealing attacks on voting systems. Selective audits of machines' vote counts can identify evidence of tampering, and Vorobeychik says the best election districts to audit are those that might be the most enticing targets.
Game theory is invaluable for both adopting an attacker's mindset and gauging the likelihood of being detected by auditors, Vorobeychik says. "If we randomly pick which districts to audit, the added unpredictability makes the attacker's choice considerably more difficult, and less certain of success," he says. "What we end up with is an audit plan that is admittedly challenging to compute, but also imposes high calculation burdens on an attacker seeking to evade detection and reduces or eliminates the impact of nearly all attacks."
From The Conversation
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA
No entries found