Guest Lecture: Privacy-Preserving Public Information for Sequential Games
نویسنده
چکیده
Suppose a collection of strategic agents are trying to make some decision. For example, a collection of investment banks are trying to decide how to invest their money. The payoffs of the players will depend upon the actions of the other players; as more people choose a particular investment, the value to investers who are “later to the game” will be smaller than the value for earlier investors. If players make these decisions in some order, and have perfect information about earlier players’ decisions, one can analyze the social welfare of various strategies or equilibrium concepts. In particular, if the utility of a player does not depend on later players, only upon the decisions made by earlier players, the greedy strategy is a dominant strategy (and the only undominated strategy). On the other hand, banks may not want their investment decisions to be public knowledge for their competition, which motivates looking into what happens if players are only provided approximate, private information regarding previous players’ actions. In our work, we ask how well different strategies perform with respect to this approximate information. In particular, today I’ll talk about how one can show that players playing greedily respect to approximate information gives social welfare whose ratio with the social wefare of OPT is bounded. I’ll start by describe a classic result from online optimization, and show how it extends to the setting where the information provided to distributed optimizers is only approximate. I’ll then show an improved privacy-preserving counter scheme which gives better approximation guarantees in our settings, by combining additive and multiplicative approximation guarantees.
منابع مشابه
X Privacy-Preserving Public Information for Sequential Games
In settings with incomplete information, players can find it difficult to coordinate to find states with good social welfare. For example, in financial settings, if a collection of financial firms have limited information about each other’s strategies, some large number of them may choose the same high-risk investment in hopes of high returns. While this might be acceptable in some cases, the e...
متن کاملA centralized privacy-preserving framework for online social networks
There are some critical privacy concerns in the current online social networks (OSNs). Users' information is disclosed to different entities that they were not supposed to access. Furthermore, the notion of friendship is inadequate in OSNs since the degree of social relationships between users dynamically changes over the time. Additionally, users may define similar privacy settings for their f...
متن کاملAdaptive Data Analysis
These lecture notes are based on [BNS+16] and were compiled for a guest lecture in the course CS229r “Information Theory in Computer Science” taught by Madhu Sudan at Harvard University in Spring 2016. Menu for today’s lecture: • Motivation • Model • Overfitting & comparison to non-adaptive data analysis • What can we do adaptively? • KL Divergence recap • Proof • Differential privacy (time per...
متن کاملA Lightweight Privacy-preserving Authenticated Key Exchange Scheme for Smart Grid Communications
Smart grid concept is introduced to modify the power grid by utilizing new information and communication technology. Smart grid needs live power consumption monitoring to provide required services and for this issue, bi-directional communication is essential. Security and privacy are the most important requirements that should be provided in the communication. Because of the complex design of s...
متن کاملارایه یک روش جدید انتشار دادهها با حفظ محرمانگی با هدف بهبود دقّت طبقهبندی روی دادههای گمنام
Data collection and storage has been facilitated by the growth in electronic services, and has led to recording vast amounts of personal information in public and private organizations databases. These records often include sensitive personal information (such as income and diseases) and must be covered from others access. But in some cases, mining the data and extraction of knowledge from thes...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2014