Abstract: In some embodiments, a method includes receiving, at a host device, a signal indicative of an authentication request for a client device to access a resource from a set of resources. A resource confidence value associated with the authentication request is calculated based at least in part on (1) a threat confidence vector associated with at least one risk mitigation score for each threat from a set of threats and (2) a set of resource vulnerability scores associated with the resource and each threat from the set of threats. The resource confidence value is compared to a resource confidence criterion associated with the resource from the set of resources. A signal indicative of a positive authentication is sent from the host device to the client device when the resource confidence value satisfies the resource confidence criterion such that the client device is granted access to the resource.

Abstract: A device for providing access to a remote site is disclosed. Access to the device is gained through an authentication process during which a user password and biometrics are provided to the device. The device also includes a security feature such that only authorized users of the specific device can gain access to it. Once authenticated, the device authorizes access to a remote site (e.g., a web site or a server on a local area network). The communications from the device to the remote site is encrypted and further the hand-held device uses a computer generated password to gain access to the site. In this way, user generated passwords, which are typically simple and infrequently changed, are avoided in favor of a more complex and frequently changed computer generated passwords for site access.

Abstract: A device for providing access to a remote site is disclosed. Access to the device is gained through an authentication process during which a user password and biometrics are provided to the device. The device also includes a security feature such that only authorized users of the specific device can gain access to it. Once authenticated, the device authorizes access to a remote site (e.g., a web site or a server on a local area network). The communications from the device to the remote site is encrypted and further the hand-held device uses a computer generated password to gain access to the site. In this way, user generated passwords, which are typically simple and infrequently changed, are avoided in favor of a more complex and frequently changed computer generated passwords for site access.

Abstract: A new algorithm for reinforcement learning, advantage updating, is proposed. Advantage updating is a direct learning technique; it does not require a model to be given or learned. It is incremental, requiring only a constant amount of calculation per time step, independent of the number of possible actions, possible outcomes from a given action, or number of states. Analysis and simulation indicate that advantage updating is applicable to reinforcement learning systems working in continuous time (or discrete time with small time steps) for which Q-learning is not applicable. Simulation results are presented indicating that for a simple linear quadratic regulator (LQR) problem with no noise and large time steps, advantage updating learns slightly faster than Q-learning. When there is noise or small time steps, advantage updating learns more quickly than Q-learning by a factor of more than 100,000. Convergence properties and implementation issues are discussed.