arXiv subjects

Dimitri Bertsekas

Publications and source records attributed to Dimitri Bertsekas.

27 records · Page 2Linked to original sources

arXiv2020

Multiagent Rollout and Policy Iteration for POMDP with Application to Multi-Robot Repair Problems

In this paper we consider infinite horizon discounted dynamic programming problems with finite state and control spaces, partial state observations, and a multiagent structure. We discuss and compare algorithms that simultaneously or sequentially optimize the agents' controls by using multistep lookahead, truncated rollout with a known base policy, and a terminal cost function approximation. Our methods specifically address the computational challenges of partially observable multiagent problems. In particular: 1) We consider rollout algorithms that dramatically reduce required computation while preserving the key cost improvement property of the standard rollout method. The per-step computational requirements for our methods are on the order of $O(Cm)$ as compared with $O(C^m)$ for standard rollout, where $C$ is the maximum cardinality of the constraint set for the control component of each agent, and $m$ is the number of agents. 2) We show that our methods can be applied to challenging problems with a graph structure, including a class of robot repair problems whereby multiple robots collaboratively inspect and repair a system under partial information. 3) We provide a simulation study that compares our methods with existing methods, and demonstrate that our methods can handle larger and more complex partially observable multiagent problems (state space size $10^{37}$ and control space size $10^{7}$, respectively). Finally, we incorporate our multiagent rollout algorithms as building blocks in an approximate policy iteration scheme, where successive rollout policies are approximated by using neural network classifiers. While this scheme requires a strictly off-line implementation, it works well in our computational experiments and produces additional significant performance improvement over the single online rollout iteration method.

cs.RO↗

arXiv2020

Multiagent Value Iteration Algorithms in Dynamic Programming and Reinforcement Learning

We consider infinite horizon dynamic programming problems, where the control at each stage consists of several distinct decisions, each one made by one of several agents. In an earlier work we introduced a policy iteration algorithm, where the policy improvement is done one-agent-at-a-time in a given order, with knowledge of the choices of the preceding agents in the order. As a result, the amount of computation for each policy improvement grows linearly with the number of agents, as opposed to exponentially for the standard all-agents-at-once method. For the case of a finite-state discounted problem, we showed convergence to an agent-by-agent optimal policy. In this paper, this result is extended to value iteration and optimistic versions of policy iteration, as well as to more general DP problems where the Bellman operator is a contraction mapping, such as stochastic shortest path problems with all policies being proper.

math.OC↗

arXiv2020

Constrained Multiagent Rollout and Multidimensional Assignment with the Auction Algorithm

We consider an extension of the rollout algorithm that applies to constrained deterministic dynamic programming, including challenging combinatorial optimization problems. The algorithm relies on a suboptimal policy, called base heuristic. Under suitable assumptions, we show that if the base heuristic produces a feasible solution, the rollout algorithm has a cost improvement property: it produces a feasible solution, whose cost is no worse than the base heuristic's cost. We then focus on multiagent problems, where the control at each stage consists of multiple components (one per agent), which are coupled either through the cost function or the constraints or both. We show that the cost improvement property is maintained with an alternative implementation that has greatly reduced computational requirements, and makes possible the use of rollout in problems with many agents. We demonstrate this alternative algorithm by applying it to layered graph problems that involve both a spatial and a temporal structure. We consider in some detail a prominent example of such problems: multidimensional assignment, where we use the auction algorithm for 2-dimensional assignment as a base heuristic. This auction algorithm is particularly well-suited for our context, because through the use of prices, it can advantageously use the solution of an assignment problem as a starting point for solving other related assignment problems, and this can greatly speed up the execution of the rollout algorithm.

math.OC↗

arXiv2020

Multiagent Rollout Algorithms and Reinforcement Learning

We consider finite and infinite horizon dynamic programming problems, where the control at each stage consists of several distinct decisions, each one made by one of several agents. We introduce an approach, whereby at every stage, each agent's decision is made by executing a local rollout algorithm that uses a base policy, together with some coordinating information from the other agents. The amount of local computation required at every stage by each agent is independent of the number of agents, while the amount of total computation (over all agents) grows linearly with the number of agents. By contrast, with the standard rollout algorithm, the amount of total computation grows exponentially with the number of agents. Despite the drastic reduction in required computation, we show that our algorithm has the fundamental cost improvement property of rollout: an improved performance relative to the base policy. We also discuss possibilities to improve further the method's computational efficiency through limited agent coordination and parallelization of the agents' computations. Finally, we explore related approximate policy iteration algorithms for infinite horizon problems, and we prove that the cost improvement property steers the algorithm towards convergence to an agent-by-agent optimal policy.

cs.LG↗

arXiv2020

Reinforcement Learning for POMDP: Partitioned Rollout and Policy Iteration with Application to Autonomous Sequential Repair Problems

In this paper we consider infinite horizon discounted dynamic programming problems with finite state and control spaces, and partial state observations. We discuss an algorithm that uses multistep lookahead, truncated rollout with a known base policy, and a terminal cost function approximation. This algorithm is also used for policy improvement in an approximate policy iteration scheme, where successive policies are approximated by using a neural network classifier. A novel feature of our approach is that it is well suited for distributed computation through an extended belief space formulation and the use of a partitioned architecture, which is trained with multiple neural networks. We apply our methods in simulation to a class of sequential repair problems where a robot inspects and repairs a pipeline with potentially several rupture sites under partial information about the state of the pipeline.

cs.RO↗

arXiv2019

Biased Aggregation, Rollout, and Enhanced Policy Improvement for Reinforcement Learning

We propose a new aggregation framework for approximate dynamic programming, which provides a connection with rollout algorithms, approximate policy iteration, and other single and multistep lookahead methods. The central novel characteristic is the use of a bias function $V$ of the state, which biases the values of the aggregate cost function towards their correct levels. The classical aggregation framework is obtained when $V\equiv0$, but our scheme works best when $V$ is a known reasonably good approximation to the optimal cost function $J^*$. When $V$ is equal to the cost function $J_μ$ of some known policy $μ$ and there is only one aggregate state, our scheme is equivalent to the rollout algorithm based on $μ$ (i.e., the result of a single policy improvement starting with the policy $μ$). When $V=J_μ$ and there are multiple aggregate states, our aggregation approach can be used as a more powerful form of improvement of $μ$. Thus, when combined with an approximate policy evaluation scheme, our approach can form the basis for a new and enhanced form of approximate policy iteration. When $V$ is a generic bias function, our scheme is equivalent to approximation in value space with lookahead function equal to $V$ plus a local correction within each aggregate state. The local correction levels are obtained by solving a low-dimensional aggregate DP problem, yielding an arbitrarily close approximation to $J^*$, when the number of aggregate states is sufficiently large. Except for the bias function, the aggregate DP problem is similar to the one of the classical aggregation framework, and its algorithmic solution by simulation or other methods is nearly identical to one for classical aggregation, assuming values of $V$ are available when needed.

cs.LG↗

arXiv2017

Affine Monotonic and Risk-Sensitive Models in Dynamic Programming

In this paper we consider a broad class of infinite horizon discrete-time optimal control models that involve a nonnegative cost function and an affine mapping in their dynamic programming equation. They include as special cases classical models such as stochastic undiscounted nonnegative cost problems, stochastic multiplicative cost problems, and risk-sensitive problems with exponential cost. We focus on the case where the state space is finite and the control space has some compactness properties. We assume that the affine mapping has a semicontractive character, whereby for some policies it is a contraction, while for others it is not. In one line of analysis, we impose assumptions that guarantee that the latter policies cannot be optimal. Under these assumptions, we prove strong results that resemble those for discounted Markovian decision problems, such as the uniqueness of solution of Bellman's equation, and the validity of forms of value and policy iteration. In the absence of these assumptions, the results are weaker and unusual in character: the optimal cost function need not be a solution of Bellman's equation, and an optimal policy may not be found by value or policy iteration. Instead the optimal cost function over just the contractive policies solves Bellman's equation, and can be computed by a variety of algorithms.

math.OC↗

arXiv2016

Robust Shortest Path Planning and Semicontractive Dynamic Programming

In this paper we consider shortest path problems in a directed graph where the transitions between nodes are subject to uncertainty. We use a minimax formulation, where the objective is to guarantee that a special destination state is reached with a minimum cost path under the worst possible instance of the uncertainty. Problems of this type arise, among others, in planning and pursuit-evasion contexts, and in model predictive control. Our analysis makes use of the recently developed theory of abstract semicontractive dynamic programming models. We investigate questions of existence and uniqueness of solution of the optimality equation, existence of optimal paths, and the validity of various algorithms patterned after the classical methods of value and policy iteration, as well as a Dijkstra-like algorithm for problems with nonnegative arc lengths.

cs.DS↗

arXiv2012

Discretized Approximations for POMDP with Average Cost

In this paper, we propose a new lower approximation scheme for POMDP with discounted and average cost criterion. The approximating functions are determined by their values at a finite number of belief points, and can be computed efficiently using value iteration algorithms for finite-state MDP. While for discounted problems several lower approximation schemes have been proposed earlier, ours seems the first of its kind for average cost problems. We focus primarily on the average cost case, and we show that the corresponding approximation can be computed efficiently using multi-chain algorithms for finite-state MDP. We give a preliminary analysis showing that regardless of the existence of the optimal average cost J in the POMDP, the approximation obtained is a lower bound of the liminf optimal average cost function, and can also be used to calculate an upper bound on the limsup optimal average cost function, as well as bounds on the cost of executing the stationary policy associated with the approximation. Weshow the convergence of the cost approximation, when the optimal average cost is constant and the optimal differential cost is continuous.

cs.AI↗