We consider model-based reinforcement learning in finite Markov Decision Processes (MDPs), focussing on so-called optimistic strategies. Optimism is usually implemented by carryin...
In this paper, we present a novel approach to enhance hierarchical phrase-based machine translation systems with linguistically motivated syntactic features. Rather than directly ...
This paper presents a cascaded scheme with block-based and pixel-based codebooks for background subtraction. The codebook is mainly used to compress information to achieve high ef...
- This paper demonstrates that simple yet important characteristics of coevolution can occur in evolutionary algorithms when only a few conditions are met. We find that interaction...
For a reproduced sound field, the competing goals between the listening area and reproduction accuracy in an actual environment is one of the most important problems in sound fi...