Decision trees have proved to be valuable tools for the description, classification and generalization of data. Work on constructing decision trees from data exists in multiple disciplines such as statistics, pattern recognition, decision theory, signal processing, machine learning and artificial neural networks. Researchers in these disciplines, sometimes working on quite different problems, identified similar issues and heuristics for decision tree construction. This paper surveys existing work on decision tree construction, attempting to identify the important issues involved, directions the work has taken and the current state of the art. Keywords: classification, tree-structured classifiers, data compaction 1. Introduction Advances in data collection methods, storage and processing technology are providing a unique challenge and opportunity for automated data exploration techniques. Enormous amounts of data are being collected daily from major scientific projects e.g., Human Genome...

Classic heuristic search algorithms can find solutions that take the form of a simple path (A*), a tree, or an acyclic graph (AO*). In this paper, we describe a novel generalization of heuristic search, called LAO*, that can find solutions with loops. We show that LAO* can be used to solve Markov decision problems and that it shares the advantage heuristic search has over dynamic programming for other classes of problems. Given a start state, it can find an optimal solution without evaluating the entire state space. 2001 Elsevier Science B.V. All rights reserved. Keywords: Heuristic search; Dynamic programming; Markov decision problems 1.

We consider the approximate solution of discrete optimization problems using procedures that are capable of magnifying the effectiveness of any given heuristic algorithm through sequential application. In particular, we embed the problem within a dynamic programming framework, and we introduce several types of rollout algorithms, which are related to notions of policy iteration. We provide conditions guaranteeing that the rollout algorithm improves the performance of the original heuristic algorithm. The method is illustrated in the context of a machine maintenance and repair problem.

Heuristic search algorithms can find solutions that take the form of a simple path (A*), a tree or an acyclic graph (AO*). We present a novel generalization of heuristic search (called LAO*) that can find solutions with loops, that is, solutions that take the form of a cyclic graph. We show that it can be used to solve Markov decision problems without evaluating the entire state space, giving it an advantage over dynamic-programming algorithms such as policy iteration and value iteration as an approach to stochastic planning.

In this paper, we present a comprehensive methodology for a formal, but intuitive, cause-e ect dependency modeling using multi-signal directed graphs that correspond closely to hierarchical system schematics and develop diagnostic strategies to isolate faults in the shortest possible time without making the unrealistic single fault assumption. A key feature of our methodology is that our models lend naturally to real-world necessities, such as system integration and hierarchical troubleshooting. 1

This paper studies the problem of learning diagnostic policies from training examples. A diagnostic policy is a complete description of the decision-making actions of a diagnostician (i.e., tests followed by a diagnostic decision) for all possible combinations of test results. An optimal diagnostic policy is one that minimizes the expected total cost, which isthe sum of measurement costs and misdiagnosis costs. In most diagnostic settings, there is a tradeo between these two kinds of costs. This paper formalizes diagnostic decision making as a Markov Decision Process (MDP). The paper introduces a new family of systematic search algorithms based on the AO algorithm to solve this MDP.To makeAO e cient, the paper describes an admissible heuristic that enables AO to prune large parts of the search space. The paper also introduces several greedy algorithms including some improvements over previously-published methods. The paper then addresses the question of learning diagnostic policies from examples. When the probabilities of diseases and test results are computed from training data, there is a great danger of over tting. To reduce over tting, regularizers are integrated into the search algorithms. Finally, the paper compares the proposed methods on ve benchmark diagnostic data sets. The studies show that in most cases the systematic search methods produce better diagnostic policies than the greedy methods. In addition, the studies show that for training sets of realistic size, the systematic search algorithms are practical on today's desktop computers. 1.

We study a subclass of POMDPs, called Deterministic POMDPs, that is characterized by deterministic actions and observations. These models do not provide the same generality of POMDPs yet they capture a number of interesting and challenging problems, and permit more efficient algorithms. Indeed, some of the recent work in planning is built around such assumptions mainly by the quest of amenable models more expressive than the classical deterministic models. We provide results about the fundamental properties of Deterministic POMDPs, their relation with AND/OR search problems and algorithms, and their computational complexity. 1

In this paper we present a new model for diagnosis of errors in Synchronous Sequential Circuits (SCC) on the functional level. In contrast to many previously published approaches we do not consider a specific implementation. Instead we use tests based on the transition behavior of the corresponding Finite State Machine (FSM). Thus, the approach can be used for verification and test. We describe a method for constructing a minimal cost test based on AND/OR graphs. Exact and heuristic methods are presented. First experimental results for randomly generated FSMs are given that demonstrate the efficiency of our approach. 1 Introduction Nowadays, circuit design is becoming more and more complex. Thus, the error probability also increases. Since time-to-market aspects are increasingly important it is desirable to detect errors as early as possible. Additionally, this also reduces the production costs. For this, nowadays CAD tools should also support features for error diagnosis, i.e. error...

Model-Based Diagnosis (MBD) approaches often yield a large number of diagnoses, severely limiting their practical utility. This paper presents a novel active testing approach based on MBD techniques, called FRACTAL (FRamework for ACtive Testing ALgorithms), which, given a system description, computes a sequence of control settings for reducing the number of diagnoses. The approach complements probing, sequential diagnosis, and ATPG, and applies to systems where additional tests are restricted to setting a subset of the existing system inputs while observing the existing outputs. This paper evaluates the optimality of FRACTAL, both theoretically and empirically. FRACTAL generates test vectors using a greedy, next-best strategy and a low-cost approximation of diagnostic information entropy. Further, the approximate sequence computed by FRACTAL’s greedy approach is optimal over all poly-time approximation algorithms, a fact which we confirm empirically. Extensive experimentation with ISCAS85 combinational circuits shows that FRACTAL reduces the number of remaining diagnoses according to a steep geometric decay function, even when only a fraction of inputs are available for active testing.

Abstract—Faulty components in a network need to be localized and repaired to sustain the health of the network. In this paper, we propose a novel approach that carefully combines active and passive measurements to localize faults in wireless sensor networks. More specifically, we formulate a problem of optimal sequential testing guided by end-to-end data. This problem determines an optimal testing sequence of network components based on end-to-end data in sensor networks to minimize testing cost. We prove that this problem is NP-hard and propose a greedy algorithm to solve it. Extensive simulation shows that in most settings our algorithm only requires testing a very small set of network components to localize and repair all faults in the network. Our approach is superior to using active and passive measurements in isolation. It also outperforms the state-of-theart approaches that localize and repair all faults in a network. I.