Many existing explanation methods in Bayesian networks, such as Maximum a Posteriori (MAP) assignment and Most Probable Explanation (MPE), generate complete assignments for target...
We examine the write endurance of USB flash drives using a range of approaches: chip-level measurements, reverse engineering, timing analysis, whole-device endurance testing, and ...
Partially Observable Markov Decision Processes have been studied widely as a model for decision making under uncertainty, and a number of methods have been developed to find the s...
We present a novel method for information-theoretic exploration, leveraging recent work on mapping and localization. We describe exploration as the constrained optimization proble...
The creation of life-like believable characters is emerging as the central focus of next-generation game development and is viewed as critical to obtaining true mass-market appeal...