The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Reinforcement learning (RL) can automate a wide variety of robotic skills, but learning each new skill requires considerable real-world data collection and manual representation engineering to design policy classes or features. Using deep reinforcement learning to train general purpose neural network policies alleviates some of the burden of manual representation engineering by using expressive policy...
We consider a zero-sum stochastic game where two players have a common observation of a global state, and each player makes a private observation of its local state at every time step. This asymmetry of information among the players makes it difficult to the compute the equilibrium cost (called the value of the zero-sum game). To help us determine the value of such a game, we first consider a game...
We consider a scenario in which a controller and an adversary dynamically act on a system over a finite or infinite horizon. The controller and the adversary do not want to reveal their actions to each other, and at the same time, the controller acts to minimize an expected cost, and the adversary acts to maximize it. We model this scenario as a dynamic zero-sum game, prove the existence of a unique...
Most electricity markets have multiple stages, which include one or more forward markets and the spot market. We consider two stages - a day-ahead market and a real-time market. We study equilibrium outcomes in such markets assuming demand to be deterministic. We show via counterexamples that in such two-stage electricity markets, (i) a Nash equilibrium may not exist, or (ii) there may be multiple...
We consider a three-step three-player complete information Colonel Blotto game in this paper, in which the first two players fight against a common adversary. Each player is endowed with a certain amount of resources at the beginning of the game, and the number of battlefields on which a player and the adversary fights is specified. The first two players are allowed to form a coalition if it improves...
We consider a jamming attack on a transmitter-receiver pair, in which the transmitter wants to transmit the state of an i.i.d. Gaussian process across an unsecured communication channel to the receiver while minimizing its cost functional. The transmitter decides whether or not to transmit the current state of the random process. The jammer disrupts the transmission on the channel strategically in...
We consider a model of stealthy attack on a networked control system by formulating a static zero-sum game among four players. Three of the players constitute a team of encoder, decoder and controller for a scalar discrete-time linear plant, while the fourth player is a jammer, who acts to flip the bits of the binary encoded observation signal of the communication channel between the plant and the...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.