Improving Domain-Independent Planning via Critical Section Macro-Operators

Abstract

Macro-operators, macros for short, are a well-known technique for enhancing performance of planning engines by providing “short-cuts” in the state space. Existing macro learning systems usually generate macros from most frequent sequences of actions in training plans. Such approach priorities frequently used sequences of actions over meaningful activities to be performed for solving planning tasks.
 This paper presents a technique that, inspired by resource locking in critical sections in parallel computing, learns macros capturing activities in which a limited resource (e.g., a robotic hand) is used. In particular, such macros capture the whole activity in which the resource is “locked” (e.g., the robotic hand is holding an object) and thus “bridge” states in which the resource is locked and cannot be used. We also introduce an “aggressive” variant of our technique that removes original operators superseded by macros from the domain model. Usefulness of macros is evaluated on several stateof-the-art planners, and a wide range of benchmarks from the learning tracks of the 2008 and 2011 editions of the International Planning Competition.

Planning with Critical Section Macros: Theory and Practice

“…For each individual domain, out of all considered planners, a planner which generates best quality training plans, i.e., the minimum sum of the lengths of the plans or the minimum sum of the costs of the plans (depending whether action costs are/not considered), is selected to generate training plans for that domain. This methodology, used also by BloMa (Chrpa & Siddiqui, 2015), follows an intuition that good quality training plans yield to most promising knowledge for all planners rather than when each planner generates training plans for itself, which has been empirically verified in the recent work (Chrpa & Vallati, 2019).…”

“…Although the aim of generating meaningful long macros is similar to BloMa, CSMs are targeted to bypassing states in which limited resources are being used that in consequence might help, for example, delete-relaxation-based approaches to make better heuristic estimates. CSMs can be combined with "chaining" macro generation approaches such as MUM, as shown in our previous work (Chrpa & Vallati, 2019).…”

“…Contributions of this paper, in which we significantly extend the work presented in our conference paper (Chrpa & Vallati, 2019), are as follows:…”

Reusability and Transferability of Macro Actions for Reinforcement Learning

“…The concept of macro actions has been adopted in the domain of planning [Asai and Fukunaga 2015;Botea et al 2005;Chrpa and Vallati 2019;Coles and Smith 2007;DeJong and Mooney 1986;Kaelbling 1993;Khetarpal et al 2020;Korf 1985;Newton et al 2007;Sacerdoti 1974], and has been shown to be able to provide advantages such as the embedding effect and evaluation effect [Botea et al 2005]. The former enables bypassing a series of successor states from a start state, and thus allows the search space to be changed as well as the search depth to be reduced.…”

Towards Automated Discovery of God-Like Folk Algorithms for Rubik’s Cube

“…Automated discovery of macros has proven effective in several AI planning domains (Botea et al 2005;Chrpa and Vallati 2019). This suggests a question which is the focus of this paper: whether we can automate discovery of macro databases that are both god-like and folk-like.…”