Experience replay lets online reinforcement learning agents remember and reuse experiences from the past. In prior work, experience transitions were uniformly sampled from a replay memory. However, this approach simply replays transitions at the same frequency that they were originally experienced, regardless of their significance. In this paper we develop a framework for prioritizing experienc...

Connict resolution is an important issue in knowledge representation and reasoning. A common idea of solving connicts in reasoning is to add preferences in the underlying reasoning mechanism. This paper describes extensions of Gelfond and Lifschitz's extended logic programs 5] by adding preference information. We rst propose prioritized logic programs (PLPs) in which the preference is expressed...

In this paper we present an approach to planning with prioritized goal states. To describe the preference ordering on goal states, we make use of ranked knowledge bases which induce a partial preference ordering on plans. We show how an optimal plan can be computed by assigning an integer value to each state in an appropriate manner. We also show how plan optimality can be tested in a similar f...

Message scheduling is shown to be very effective in belief propagation (BP) algorithms. However, most existing scheduling algorithms use fixed heuristics regardless of the structure of the graphs or properties of the distribution. On the other hand, designing different scheduling heuristics for all graph structures are not feasible. In this paper, we propose a reinforcement learning based messa...

A video streaming solution with the ner granular scalable (FGS) MPEG-4 stream delivered over prioritized networks is investigated in this work. An optimal truncation strategy for constant-quality rate adaptation is rst presented by embedding the rate-distortion (R-D) information based on a piecewise linear R-D model. Then, the video stream is prioritized for di erentiated dropping and forwardin...