Tractable reasoning in artificial intelligence

In this book we studied two popular techniques for limiting complexity of reasoning in Artificial Intelligence problems. The first technique, known as language restriction, has been studied in Chapter 2, referring in particular to a form of non-monotonic reasoning known as minimal reasoning. The second technique we considered is logic-based approximation and has been studied in Chapter 3. In Chapter 4 we made a first attempt at integrating the two studies on tractable reasoning, showing how complexity analysis about minimal reasoning can be used as a tool for evaluating techniques for approximate reasoning. In the present chapter we intend to comment the results we obtained and to discuss problems that are still open.