Data-driven Generation of Policies (SpringerBriefs in Computer Science)
Format: PDF / Kindle (mobi) / ePub
This Springer Brief presents a basic algorithm that provides a correct solution to finding an optimal state change attempt, as well as an enhanced algorithm that is built on top of the well-known trie data structure. It explores correctness and algorithmic complexity results for both algorithms and experiments comparing their performance on both real-world and synthetic data. Topics addressed include optimal state change attempts, state change effectiveness, different kind of effect estimators, planning under uncertainty and experimental evaluation. These topics will help researchers analyze tabular data, even if the data contains states (of the world) and events (taken by an agent) whose effects are not well understood. Event DBs are omnipresent in the social sciences and may include diverse scenarios from political events and the state of a country to education-related actions and their effects on a school system. With a wide range of applications in computer science and the social sciences, the information in this Springer Brief is valuable for professionals and researchers dealing with tabular data, artificial intelligence and data mining. The applications are also useful for advanced-level students of computer science.
Science Press, 1988. 4. L.G. Valiant. The complexity of computing the permanent. Theoretical Computer Science, 8(2):189–201, 1979. Chapter 3 Different Kinds of Effect Estimators In this chapter we introduce several sorts of effect estimator, which yield the likelihood of a given action tuple satisfying a given goal condition G. An effect estimator essentially answers the question: “if I succeed in changing the environment in this way, what is the probability that the environment satisfies my
Author Bios Bröcheler) at the 2009 International Conference on Logic Programming (ICLP). He is currently a senior researcher at the Department of Computer Science, University of Oxford (UK). Amy Sliva: Amy Sliva is a Scientist at Charles River Analytics researching artificial intelligence models and large-scale data analytics for decision-making. Dr. Sliva was previously an Assistant Professor of Computer Science and Political Science at Northeastern University where she developed
TOSCA is simply a set of tuples from the event KB, denoted tuples.N /. Tries have a unique root node. A trie is data correct if for any leaf node N there is a unique path from the root .At r1 ; Edges1 /; : : : ; .At rk 1 ; Edgesk 1 /; N such that for all t 2 tuples.N / and 3.4 Trie-Enhanced Optimal State Change Attempts (TOSCA) 25 A1 [0,1) [1,inf ) A2 A2 [0,1) [0,1) A2 0 0 0 A3 0 0 0 S1 0 0 1 A3 A3 A3 A1 0 0 0 [1,inf ) [0,1) [1,inf ) A1 A2 A3 S1 1 0 1 0 [0,1) [0,1) A1 A2 A3 S1
SCAs that can be considered in a given state is very large. This makes planning approaches infeasible since their computational cost is intimately tied to the number of possible actions in the domain (generally assumed to be fixed at a relatively small number). In the case of MDPs, even though state aggregation techniques have been investigated to keep the number of states being considered manageable [1, 3, 6], similar techniques for action aggregation have not been developed. To conclude this
that, similar to this work, is often specified as a goal condition. In this work, we have described an approach to solve problems that at first seems quite similar to those tackled by AI planning; however the main characteristic of the problems of interest are that important assumptions made in AI planning approaches cannot be made in this case. There are, however, significant assumptions made in these works that cannot always be made, such as: 1. The number of actions available to solve the