Matrix Factorization Techniques for Recommender Systems

In this paper, Monte Carlo tree search (MCTS) is introduced for controlling the Pac-Man character in the real-time game Ms Pac-Man. MCTS is used to find an optimal path for an agent at each turn, determining the move to make based on the results of numerous randomized simulations. Several enhancements are introduced in order to adapt MCTS to the real-time domain. Ms Pac-Man is an arcade game, in which the protagonist has several goals but no conclusive terminal state. Unlike games such as Chess or Go there is no state in which the player wins the game. Instead, the game has two subgoals, 1) surviving and 2) scoring as many points as possible. Decisions must be made in a strict time constraint of 40 ms. The Pac-Man agent has to compete with a range of different ghost teams, hence limited assumptions can be made about their behavior. In order to expand the capabilities of existing MCTS agents, four enhancements are discussed: 1) a variable-depth tree; 2) simulation strategies for the ghost team and Pac-Man; 3) including long-term goals in scoring; and 4) reusing the search tree for several moves with a decay factor γ. The agent described in this paper was entered in both the 2012 World Congress on Computational Intelligence (WCCI'12, Brisbane, Qld., Australia) and the 2012 IEEE Conference on Computational Intelligence and Games (CIG'12, Granada, Spain) Pac-Man Versus Ghost Team competitions, where it achieved second and first places, respectively. In the experiments, we show that using MCTS is a viable technique for the Pac-Man agent. Moreover, the enhancements improve overall performance against four different ghost teams.

Real-Time Monte Carlo Tree Search in Ms Pac-Man

Genetic algorithms (GAs) are a class of evolutionary algorithms inspired by Darwinian natural selection. They are popular heuristic optimisation methods based on simulated genetic mechanisms, i.e., mutation, crossover, etc. and population dynamical processes such as reproduction, selection, etc. Over the last decade, the possibility to emulate a quantum computer (a computer using quantum-mechanical phenomena to perform operations on data) has led to a new class of GAs known as “Quantum Genetic Algorithms” (QGAs). In this review, we present a discussion, future potential, pros and cons of this new class of GAs. The review will be oriented towards computer scientists interested in QGAs “avoiding” the possible difficulties of quantum-mechanical phenomena.

Quantum Genetic Algorithms for Computer Scientists

Ms Pac-Man is one of the most challenging test beds in game artificial intelligence (AI). Genetic programming and Monte Carlo Tree Search (MCTS) have already been successful applied to several games including Pac-Man. In this paper, we use Monte Carlo Tree Search to create a Ms Pac-Man playing agent before using genetic programming to enhance its performance by evolving a new default policy to replace the random agent used in the simulations. The new agent with the evolved default policy was able to achieve an 18% increase on its average score over the agent with random default policy.

Using genetic programming to evolve heuristics for a Monte Carlo Tree Search Ms Pac-Man agent

Algorithms for computing strategies in two-player simultaneous move games

In this paper, we present an application of Monte Carlo tree search (MCTS) to control ghosts in the game called Ms Pac-Man Our proposed ghost team consists of a ghost controlled by rules and three ghosts controlled individually by different MCTS Given a limited time response, in order to increase the reliability of MCTS results, we introduce a mechanism for predicting Ms Pac-Man's future movements and use this mechanism for simulating Ms Pac-Man during Monte Carlo simulations Our ghost team won the first Ms Pac-Man Versus Ghost Team Competition at the 2011 IEEE Congress on Evolutionary Computation (CEC) Its performances for a variety of design choices are also shown and discussed

Monte Carlo Tree Search for Collaboration Control of Ghosts in Ms. Pac-Man

We present an application of Monte-Carlo Tree Search (MCTS) to controlling ghosts in the game of Ms Pac-Man. We approach the problem by performing MCTS on each ghost's tree that represents the game state from the ghost's perspective. Our goal is to create a strong ghost team that is adaptable to a variety of Ms Pac-Man's play styles. This ghost team (ICE gUCT) won the CEC 2011 Ms Pac-Man vs Ghost Team Competition for the ghost side.

Applying Monte-Carlo Tree Search to collaboratively controlling of a Ghost Team in Ms Pac-Man

In this paper, we present a new matrix factorization method for recommender system problems, named bounded-SVD, which utilizes the constraint that all the ratings in the rating matrix are bounded within a pre-determined range. In our proposed method, the bound constraints are included in the objective function so that both the task of minimizing errors and the constraints are taken into account during the optimization process. For evaluation, we compare the performance of bounded-SVD with an existing method, called Bounded Matrix Factorization (BMF), which also uses the bound constraints on the ratings. The results on major real-world recommender system datasets show that our method outperforms BMF in almost cases and it is also faster and more simple to implement than BMF. Moreover, the way the bound constraints are integrated in bounded-SVD can also be applied to other optimization problems with bound constraints as well.

/pdf/bounded-svd-a-matrix-factorization-method-with-bound-1cdfc8btjl.pdf

Bounded-SVD: A Matrix Factorization Method with Bound Constraints for Recommender Systems

In this paper, we present an enhanced quantum-inspired genetic algorithm (eQiGA) with a combination of proposed mechanisms: two search supportive schemes and artificial entanglement. This combination is aimed at balancing exploration and exploitation. Two schemes, namely Explore and Exploit scheme are designed with aggressive specific roles reflecting its name. Entanglement is considered to be one of the significant strengths in quantum computing aside the probabilistic representation and superposition. Hence we attempt to apply its concept as part of our strategy for its potential. In addition, two new sub-strategies are proposed: fitness threshold, and quantum side-stepping. The algorithm is tested on multiple numerical optimization functions, and significant results of improved performance are obtained, studied, and discussed.

/pdf/quantum-inspired-genetic-algorithm-with-two-search-kwlc4iejzt.pdf

Quantum-inspired genetic algorithm with two search supportive schemes and artificial entanglement

Real-Time Strategy (RTS) games typically take place in a war-like setting and are accompanied with complicated game play. They are not only difficult for human players to master, but also provide a challenging platform for AI research. In a typical RTS game, such as StarCraft, WarCraft, or Age of Empires, knowing what the opponent is doing is a great advantage and sometimes an important key to win the game. For that, good scouting is required. As subsequent work for improving the scouting agent in our StarCraft AI bot-IceBot-the winner of the mixed division in Student StarCraft AI Tournament 2012, this paper proposes a method that applies potential flows to controlling scout units in StarCraft. The proposed method outperforms an existing scouting method as well as a modified version of this existing method and is comparable to scouting by human players.

Kien Quang Nguyen

Papers

Monte Carlo Tree Search for Collaboration Control of Ghosts in Ms. Pac-Man

Applying Monte-Carlo Tree Search to collaboratively controlling of a Ghost Team in Ms Pac-Man

Bounded-SVD: A Matrix Factorization Method with Bound Constraints for Recommender Systems

Quantum-inspired genetic algorithm with two search supportive schemes and artificial entanglement

Potential flows for controlling scout units in StarCraft