Explain Reinforcement Learning Agents Through Fuzzy Rule Reconstruction

被引:0
作者
Ou, Liang [1 ]
Chang, Yu-Chen [1 ]
Wang, Yu-Kai [1 ]
Lin, Chin-Teng [1 ]
机构
[1] Univ Technol Sydney, Sch Comp Sci, Sydney, NSW, Australia
来源
2023 IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS, FUZZ | 2023年
基金
澳大利亚研究理事会;
关键词
Explainable AI; Reinforcement Learning; Fuzzy Neural Network; Generative Model;
D O I
10.1109/FUZZ52849.2023.10309670
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
There has been a lot of interest in making reinforcement learning (RL) models more explainable, as their applications become more widespread. Currently, most of the explainable RL (xRL) models focus on improving the transparency of the agent's observations, rather than the relationship between the agent's states and actions. This study introduces the Explainable Fuzzy Reconstruction Net (EFRN), which aims to interpret these relationships in RL. The EFRN utilizes the interpretability of Fuzzy Neural Networks (FNNs) to create IF-THEN rules and a generative model to showcase the learned knowledge. The IF-THEN rules can be expressed in a way that is easily understandable for humans, such as "IF A THEN B." The generative model then visualizes the state as patterns that is easy for humans to comprehend. The results of the study shows that the proposed EFRN maintains the same level of performance as traditional RL methods and significantly improves the explainability of the RL agents both globally and locally.
引用
收藏
页数:6
相关论文
共 19 条
  • [1] Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
    Adadi, Amina
    Berrada, Mohammed
    [J]. IEEE ACCESS, 2018, 6 : 52138 - 52160
  • [2] FWNNet: Presentation of a New Classifier of Brain Tumor Diagnosis Based on Fuzzy Logic and the Wavelet-Based Neural Network Using Machine-Learning Methods
    Ahmadi, Mohsen
    Ahangar, Fatemeh Dashti
    Astaraki, Nikoo
    Abbasi, Mohammad
    Babaei, Behzad
    [J]. COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE, 2021, 2021
  • [3] Alonso Moral J. M., 2021, Explainable Fuzzy Systems: Paving the Way from Interpretable Fuzzy Systems to Explainable AI Systems, P219
  • [4] A Time Series Based Explainable Interval Type Fuzzy Logic System
    Bhatia, Ashish
    Hagras, Hani
    [J]. 2022 IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS (FUZZ-IEEE), 2022,
  • [5] Borowski J, 2021, Arxiv, DOI arXiv:2010.12606
  • [6] COREYES JD, 2018, PR MACH LEARN RES, V80, pNI107
  • [7] Dahalan A. J., 2021, IAES International Journal of Artificial Intelligence, V10, P1036
  • [8] Hayat A, 2019, Arxiv, DOI arXiv:1905.10404
  • [9] Coordinated control system modeling of ultra-supercritical unit based on a new fuzzy neural network
    Hou, Guolian
    Xiong, Jian
    Zhou, Guiping
    Gong, Linjuan
    Huang, Congzhi
    Wang, Shunjiang
    [J]. ENERGY, 2021, 234
  • [10] Jeyakumar JV, 2020, ADV NEUR IN, V33