Given a class of students, and a pool of questions in the domain of study, what subset will constitute a good exam? Millions of educators are dealing with this difficult problem worldwide, yet exams are still composed manually in non-systematic ways. In this work we present a novel algorithmic framework for exam composition. Our framework requires two input components: a student population represented by a distribution over overlay models, each consisting of a set of mastered abilities, or actions; and a target model ordering that, given any two student models, defines which should be given the higher grade. To determine the performance of a student model on a potential question, we test whether it satisfies a disjunctive action landmark, i.e., whether its abilities are sufficient to follow at least one solution path. We present a novel utility function for evaluating exams, using the described components. An exam is highly evaluated if it is expected to order the student population with high correlation to the target order.The merit of our algorithmic framework is exemplified with real auto-generated questions in the domain of middle-school algebra.

Computing the set of optimal solutions for a multi-objective constraint optimisation problem can be computationally very challenging. Also, when solutions are only partially ordered, there can be a number of different natural notions of optimality, one of the most important being the notion of Possibly Optimal, i.e., optimal in at least one scenario compatible with the inter-objective tradeoffs. We develop an AND/OR Branch-and-Bound algorithm for computing the set of Possibly Optimal solutions, and compare variants of the algorithm experimentally.

We propose a method to guide a Monte Carlo search in the initial moves of the game of Go. Our method matches the current state of a Go board against clusters of board configurations that are derived from a large number of games played by experts. The main advantage of this method is that it does not require an exact match of the current board, and hence is effective for a longer sequence of moves compared to traditional opening books. We apply this method to two different open-source Go-playing programs. Our experiments show that this method, through its filtering or biasing the choice of a next move to a small subset of possible moves, improves play effectively in the initial moves of a game.

Recently, a number of papers have shown that for many domains, using multiple heuristics in independent searches performs better than combining them into a single heuristic. Furthermore, using a large number of “weak” heuristics could potentially eliminate the need for the careful design of a few. The standard approach to distribute computation in these multi-heuristic searches is to rotate through the heuristics in a round-robin fashion. However, this strategy can be inefficient especially in the case when only a few of the heuristics are leading to progress. In this paper, we present two principled methods to adaptively distribute computation time among the different searches of the Multi- Heuristic A* algorithm. The first method, Meta-A*, constructs and searches a meta-graph, which represents the problem of finding the best heuristic as the problem of minimizing the total number of expansions. The second treats the scheduling of searches with different heuristics as a multi-armed bandit problem. It applies Dynamic Thompson Sampling (DTS) to keep track of what searches are making progress the most and continuously re-computes the schedule of searches based on this information. We provide a theoretical analysis and compare our new strategies with the round-robin method on a 12-DOF full-body motion planning problem and on sliding tile puzzle problems. In these experiments, we used up to 20 heuristics and observed a several times speedup without loss in solution quality.

Several applications including text-to-speech require some normalized format of non-standard words in various domains such as numbers, dates, and currencies and in various human languages. The traditional approach of manually constructing a program for such a normalization task requires expertise in both programming and target (human) language and further does not scale to a large number of domain, format, and target language combinations. We propose to learn programs for such normalization tasks through examples. We present a domain-specific programming language that offers appropriate abstractions for succinctly describing such normalization tasks, and then present a novel search algorithm that can effectively learn programs in this language from input-output examples. We also briefly describe domain-specific heuristics for guiding users of our system to provide representative examples for normalization tasks related to that domain. Our experiments show that weare able to effectively learn desired programs for a variety of normalization tasks.

Planning an interplanetary trajectory is a very complex task, traditionally accomplished by domain experts using computer-aided design tools. Recent advances in trajectory optimization allow automation of part of the trajectory design but have yet to provide an efficient way to select promising planetary encounter sequences. In this work, we present a heuristic-free approach to automated trajectory planning (including the encounter sequence planning) based on Monte Carlo Tree Search (MCTS). We discuss a number of modifications to traditional MCTS unique to the domain of interplanetary trajectory planning and provide results on the Rosetta and Cassini-Huygens interplanetary mission design problems. The resulting heuristic-free method is found to be orders of magnitude more efficient with respect to a standard tree search with heuristic-based pruning which is the current state-of-the art in this domain.

Monte Carlo Tree Search (MCTS) is the state of the art algorithm for many games including the game of Go and General Game Playing (GGP). The standard algorithm for MCTS is Upper Confidence bounds applied to Trees (UCT). For games such as Go a big improvement over UCT is the Rapid Action Value Estimation (RAVE) heuristic. We propose to generalize the RAVE heuristic so as to have more accurate estimates near the leaves. We test the resulting algorithm named GRAVE for Atarigo, Knighthrough, Domineering and Go.

The problem of finding a minimum vertex cover (MinVC) in a graph is a well known NP-hard problem with important applications. There has been much interest in developing heuristic algorithms for finding a "good" vertex cover in graphs. In practice, most heuristic algorithms for MinVC are based on the local search method. Previously, local search algorithms for MinVC have focused on solving academic benchmarks where the graphs are of relatively small size, and they are not suitable for solving massive graphs as they usually have high-complexity heuristics. In this paper, we propose a simple and fast local search algorithms called FastVC for solving MinVC in massive graphs, which is based on two low-complexity heuristics. Experimental results on a broad range of real world massive graphs show that FastVC finds much better vertex cover (and thus also independent sets) than state of the art local search algorithms for MinVC.

Automatic algorithm configurators are important practical tools for improving program performance measures, such as solution time or prediction accuracy. Local search approaches in particular have proven very effective for tuning algorithms. In sequential local search, the use of predictive models has proven beneficial for obtaining good tuning results. We study the use of non-parametric models in the context of population-based algorithm configurators. We introduce a new model designed specifically for the task of predicting high-performance regions in the parameter space. Moreover, we introduce the ideas of genetic engineering of offspring as well as sexual selection of parents. Numerical results show that model-based genetic algorithms significantly improve our ability to effectively configure algorithms automatically.

Marginal MAP is known to be a difficult task for graphical models, particularly because the evaluation of each MAP assignment involves a conditional likelihood computation. In order to minimize the number of likelihood evaluations, we focus in this paper on best-first search strategies for exploring the space of partial MAP assignments. We analyze the potential relative benefits of several best-first search algorithms and demonstrate their effectiveness against recent branch and bound schemes through extensive empirical evaluations. Our results show that best-first search improves significantly over existing depth-first approaches, in many cases by several orders of magnitude, especially when guided by relatively weak heuristics.