In this paper, we present a rapid learning algorithm called Dyna-QPC. The proposed algorithm requires considerably less training time than Q-learning and Table-based Dyna-Q algorithm, making it applicable to real-world control tasks. The Dyna-QPC algorithm is a combination of existing learning techniques: CMAC, Q-learning, and prioritized sweeping. In a practical experiment, the Dyna-QPC algori...

Planning and trajectory optimization can readily be used for kinematic control of robotic manipulation. However, planning dynamic motor skills requires a detailed physical simulation, and some aspects of the task, such as contacts, are very difficult to simulate with enough accuracy for dynamic manipulation. Alternatively, manipulation skills can be learned from experience, allowing them to def...

Reinforcement Learning algorithms can learn complex behavioral patterns for sequential decision making tasks wherein an agent interacts with an environment and acquires feedback in the form of rewards sampled from it. Traditionally, such algorithms make decisions, i.e., select actions to execute, at every single time step of the agent-environment interactions. In this paper, we propose a novel ...

Reinforcement Learning algorithms can learn complex behavioral patterns for sequential decision making tasks wherein an agent interacts with an environment and acquires feedback in the form of rewards sampled from it. Traditionally, such algorithms make decisions, i.e., select actions to execute, at every single time step of the agent-environment interactions. In this paper, we propose a novel ...

Employing in-depth, elite interviews, this empirical research contributes to understanding the dynamics among policy windows, policy change, and organizational learning. First, although much of the research on agenda setting-how issues attract enough attention that action is taken to address them-has been conducted at the national scale, this work explores the subnational, regional scale. With ...

Medical travel has expanded rapidly in recent years, resulting in new markets and increased access to medical care. Whereas several studies investigated the motives of individuals seeking healthcare abroad, the conventional analytical approach is limited by substantial caveats. Classical techniques as found in the literature cannot provide sufficient insight due to the nested nature of data gen...

The structure, conduct, and performance of Chinese telecommunications are analyzed to determine reform's impact on policy objectives. Recent performance suggests that a more comprehensive reform is needed. However, China's fragmented policy environment makes large-scale reform difficult. Further deregulation of the telecommunication industry requires several related initiatives: reduction of en...

When scientific research collides with social values, science's right to self-governance becomes an issue of paramount concern. In this article, I develop an account of scientific autonomy within a framework of public oversight. I argue that scientific autonomy is justified because it promotes the progress of science, which benefits society, but that restrictions on autonomy can also be justifi...

In order to influence global policy effectively, conservation scientists need to be able to provide robust predictions of the impact of alternative policies on biodiversity and measure progress towards goals using reliable indicators. We present a framework for using biodiversity indicators predictively to inform policy choices at a global level. The approach is illustrated with two case studie...