Learning a Go Heuristic with TILDE
نویسندگان
چکیده
In Go, an important factor that hinders search is the large branching factor, even in local problems. Human players are strong at recognizing frequently occurring shapes and vital points. This allows them to select the most promising moves and to prune the search tree. In this paper we argue that many of these shapes can be represented as relational concepts. We present an application of the relational learner T ILDE in which we learn a heuristic that gives values to candidate-moves in tsume-go (life and death) problems. Such a heuristic can be used to limit the number of evaluated moves. Even if all moves are evaluated, alpha-beta search can be sped up considerably when the candidate-moves are approximately ordered from good to bad. We validate our approach with experiments and analysis.
منابع مشابه
Top-down induction of logical decision trees
Top-down induction of decison trees (TDIDT) is a very popular machine learning technique. Up till now, it has mainly used for propositional learning, but seldomly for relational learning or inductive logic programming. The main contribution of this paper is the introduction of logic decision trees, which make it possible to use TDIDT in inductive logic programming. An implementation of this top...
متن کاملA multiagent architecture for a web-based adaptive educational system
The widespread use of the Web in distance learning, the particular nature of the distance learning student and the dispersion of the relevant information sources increase the importance of developing interactive systems, which adapt to the information and communication needs of each student. We have developed a multiagent decision system which adapts to the user’s needs. To best accomplish this...
متن کاملCompetitive Reinforcement Learning for Combinatorial Problems
This paper shows that the competitive learning rule found in Learning Vector Quantization (LVQ) serves as a promising function approximator to enable reinforcement learning methods to cope with a large decision search space, defined in terms of different classes of input patterns, like those found in the game of Go. In particular, this paper describes S[arsa]LVQ, a novel reinforcement learning ...
متن کاملUsing Logical Decision Trees for Clustering
A novel rst order clustering system, called C 0.5, is presented. It inherits its logical decision tree formalism from the TILDE system, but instead of using class information to guide the search, it employs the principles of instance based learning in order to perform clustering. Various experiments are discussed, which show the promise of the approach.
متن کاملJason Intentional Learning: An Operational Semantics
This paper introduces an operational semantics for defining Intentional Learning on Jason, the well known Java-based implementation of AgentSpeak(L). This semantics enables Jason to define agents capable of learning the reasons for adopting intentions based on their own experience. In this work, the use of the term Intentional Learning is strictly circumscribed to the practical rationality theo...
متن کامل