From Bandits to Monte-Carlo Tree Search: The Optimistic Principle Applied to Optimization and Planning

Abstract

This work covers several aspects of the optimism in the face of uncertainty principle applied to large scale optimization problems under finite numerical budget. The initial motivation for the research reported here originated from the empirical success of the so-called Monte-Carlo Tree Search method popularized in Computer Go and further extended to many other games as well as optimization and planning problems. Our objective is to contribute to the development of theoretical foundations of the field by characterizing the complexity of the underlying optimization problems and designing efficient algorithms with performance guarantees. The main idea presented here is that it is possible to decompose a complex decision making problem (such as an optimization problem in a large search space) into a sequence of elementary decisions, where each decision of the sequence is solved using a (stochastic) multi-armed bandit (simple mathematical model for decision making in stochastic environments). This so-called hierarchical bandit approach (where the reward observed by a bandit in the hierarchy is itself the return of another bandit at a deeper level) possesses the nice feature of starting the exploration by a quasi-uniform sampling of the space and then focusing progressively on the most promising area, at different scales, according to the evaluations observed so far, and eventually performing a local search around the global optima of the function. The performance of the method is assessed in terms of the optimality of the returned solution as a function of the number of function evaluations. Our main contribution to the field of function optimization is a class of hierarchical optimistic algorithms designed for general search spaces (such as metric spaces, trees, graphs, Euclidean spaces, ...) with different algorithmic instantiations depending on whether the evaluations are noisy or noiseless and whether some measure of the "smoothness" of the function is known or unknown. The performance of the algorithms depends on the local behavior of the function around its global optima expressed in terms of the quantity of near-optimal states measured with some metric. If this local smoothness of the function is known then one can design very efficient optimization algorithms (with convergence rate independent of the space dimension), and when it is not known, we can build adaptive techniques that can, in some cases, perform almost as well as when it is known. In order to be self-contained, we start with a brief introduction to the stochastic multi-armed bandit problem in Chapter 1 and describe the UCB (Upper Confidence Bound) strategy and several extensions. In Chapter 2 we present the Monte-Carlo Tree Search method applied to Computer Go and show the limitations of previous algorithms such as UCT (UCB applied to Trees). This provides motivation for designing theoretically well-founded optimistic optimization algorithms. The main contributions on hierarchical optimistic optimization are described in Chapters 3 and 4...

Pursuit-evasion games: a tractable framework for antijamming games in aerial attacks

“…We consider that the constants are k 1 = 1, k 2 = 500 , and k 3 = 1. The non-convex algorithm Simultaneous Optimistic Optimization (SOO) details can be found in [14,15]. This algorithm is used in order to obtain the game solution-i.e., final heading angle, which is the control, and time of capture, which is the payoff of the game.…”

“…The non-convex optimization algorithm implementation used [14,15] in the optimization and hybrid methods stops when a fixed number of iterations have been done, regardless of whether a solution was found or not. In order to study how the iteration number affects to solution obtaining, we run the algorithm three times for optimization method (using {10 3 , 10 4 , 10 5 } iterations) and for the hybrid approach (using {10 2 , 10 3 , 10 4 } iterations).…”

Large Scale Hard Sample Mining with Monte Carlo Tree Search

“…Another example of successful use of MCTS is the optimization of a "black-box" function [27,8] where the goal is to get a good estimate of the maximum of the function (deterministic or stochastic) by evaluating it only a limited number of times. The idea is to design a sequence of input samples on which the function should be evaluated given the previously observed values.…”

Planning in Discrete and Continuous Markov Decision Processes by Probabilistic Programming

“…Instead of building the full tree, UCT chooses the action a that maximizes an upper confidence bound of Q(s, a), following the principle of optimism in the face of uncertainty. Several improvements and extensions for UCT have been proposed, including handling continuous actions [13] (see [16] for a review), and continuous states [1] with a simple Gaussian distance metric; however the knowledge of the probabilistic model is not directly exploited. For continuous states, parametric function approximation is often used (e.g., linear regression), nonetheless the model needs to be carefully tailored for the domain to solve [32].…”