Semantic Memory Modeling and Memory Interaction in Learning Agents

被引:14
作者
Wang, Wenwen [1 ]
Tan, Ah-Hwee [1 ]
Teow, Loo-Nin [2 ]
机构
[1] Nanyang Technol Univ, Sch Comp Engn, Singapore 639798, Singapore
[2] DSO Natl Labs, Singapore 118230, Singapore
来源
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS | 2017年 / 47卷 / 11期
关键词
Adaptive resonance theory; learning agents; memory interactions; semantic memory; REPRESENTATION;
D O I
10.1109/TSMC.2016.2531683
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Semantic memory plays a critical role in reasoning and decision making. It enables an agent to abstract useful knowledge learned from its past experience. Based on an extension of fusion adaptive resonance theory network, this paper presents a novel self-organizing memory model to represent and learn various types of semantic knowledge in a unified manner. The proposed model, called fusion adaptive resonance theory for multimemory learning, incorporates a set of neural processes, through which it may transfer knowledge and cooperate with other long-term memory systems, including episodic memory and procedural memory. Specifically, we present a generic learning process, under which various types of semantic knowledge can be consolidated and transferred from the specific experience encoded in episodic memory. We also identify and formalize two forms of memory interactions between semantic memory and procedural memory, through which more effective decision making can be achieved. We present experimental studies, wherein the proposed model is used to encode various types of semantic knowledge in different domains, including a first-person shooting game called Unreal Tournament, the Toads and Frogs puzzle, and a strategic game known as Star Craft Broodwar. Our experiments show that the proposed knowledge transfer process from episodic memory to semantic memory is able to extract useful knowledge to enhance the performance of decision making. In addition, cooperative interaction between semantic knowledge and procedural skills can lead to a significant improvement in both learning efficiency and performance of the learning agents.
引用
收藏
页码:2882 / 2895
页数:14
相关论文
共 32 条
[1]  
[Anonymous], 1983, Canadian Psychology
[2]  
[Anonymous], 1993, Rules of the Mind
[3]   The neurobiology of semantic memory [J].
Binder, Jeffrey R. ;
Desai, Rutvik H. .
TRENDS IN COGNITIVE SCIENCES, 2011, 15 (11) :527-536
[4]   A MASSIVELY PARALLEL ARCHITECTURE FOR A SELF-ORGANIZING NEURAL PATTERN-RECOGNITION MACHINE [J].
CARPENTER, GA ;
GROSSBERG, S .
COMPUTER VISION GRAPHICS AND IMAGE PROCESSING, 1987, 37 (01) :54-115
[5]   RETRIEVAL TIME FROM SEMANTIC MEMORY [J].
COLLINS, AM ;
QUILLIAN, MR .
JOURNAL OF VERBAL LEARNING AND VERBAL BEHAVIOR, 1969, 8 (02) :240-&
[6]  
Del Missier F., 2002, P 24 ANN C COGN SCI, P262
[7]  
Eden T., 2015, REINFORCEMENT LEARNI
[8]   A COMPUTATIONAL MODEL OF SEMANTIC MEMORY IMPAIRMENT - MODALITY SPECIFICITY AND EMERGENT CATEGORY SPECIFICITY [J].
FARAH, MJ ;
MCCLELLAND, JL .
JOURNAL OF EXPERIMENTAL PSYCHOLOGY-GENERAL, 1991, 120 (04) :339-357
[9]   HIPPOCAMPAL MEDIATION OF STIMULUS REPRESENTATION - A COMPUTATIONAL THEORY [J].
GLUCK, MA ;
MYERS, CE .
HIPPOCAMPUS, 1993, 3 (04) :491-516
[10]   Topics in semantic representation [J].
Griffiths, Thomas L. ;
Steyvers, Mark ;
Tenenbaum, Joshua B. .
PSYCHOLOGICAL REVIEW, 2007, 114 (02) :211-244