A Review of Symbolic, Subsymbolic and Hybrid Methods for Sequential Decision Making

• Published in: ACM computing surveys Vol. 56; no. 11; pp. 1 - 36
• Main Authors: Núñez-Molina, Carlos, Mesejo, Pablo, Fernández-Olivares, Juan
• Format: Journal Article
• Language: English
• Published: New York, NY ACM 01.11.2024; Association for Computing Machinery
• Subjects: Automated reasoning; Computing methodologies; Constraint and logic programming; Decision making; Design; Discrete space search; Dynamic programming for Markov decision processes; Game tree search; General and reference; General literature; HCI theory, concepts and models; Heuristic function construction; Human-centered computing; Knowledge representation; Knowledge representation and reasoning; Learning latent representations; Logic and verification; Logical and relational learning; Markov decision processes; Neural networks; Partially-observable Markov decision processes; Philosophical/theoretical foundations of artificial intelligence; Planning and scheduling; Reference works; Reliability; Search with partial observations; Sequential decision making; Supervised learning; Surveys and overviews; Theory of computation; Validation; Verification; Neurosymbolic AI
• ISSN: 0360-0300; 1557-7341
• DOI: 10.1145/3663366

In the field of Sequential Decision Making (SDM), two paradigms have historically vied for supremacy: Automated Planning (AP) and Reinforcement Learning (RL). In the spirit of reconciliation, this article reviews AP, RL and hybrid methods (e.g., novel learn to plan techniques) for solving Sequential Decision Processes (SDPs), focusing on their knowledge representation: symbolic, subsymbolic, or a combination. Additionally, it also covers methods for learning the SDP structure. Finally, we compare the advantages and drawbacks of the existing methods and conclude that neurosymbolic AI poses a promising approach for SDM, since it combines AP and RL with a hybrid knowledge representation.