Design and Implementation of an Explainable Bidirectional LSTM Model Based on Transition System Approach for Cooperative AI-Workers

Abstract

Recently, interest in the Cyber-Physical System (CPS) has been increasing in the manufacturing industry environment. Various manufacturing intelligence studies are being conducted to enable faster decision-making through various reliable indicators collected from the manufacturing process. Artificial intelligence (AI) and Machine Learning (ML) have advanced enough to give various possibilities of predicting manufacturing time, which can help implement CPS in manufacturing environments, but it is difficult to secure reliability because it is difficult to understand how AI works, and although it can offer good results, it is often not applied to industries. In this paper, Bidirectional Long Short Term Memory (BI-LSTM) is used to predict process execution time, which is an indicator that can be used as a basis for CPS in the manufacturing process, and the Shapley Additive Explanations (SHAP) algorithm is used to explain how artificial intelligence works. The experimental results of this paper, applying manufacturing data, prove that the results derived from SHAP are effective for workers and AI to collaborate.

Explainable AI Over the Internet of Things (IoT): Overview, State-of-the-Art and Future Directions

“…• Model-based: Model interpretation is the core aspect of any AI-based system. It ensures to provide better decision-making policies with fairness, transparency, and accountability of the results, which may give enough confidence to humans [59]. With the value and accuracy interpreted from the models, it can be easy to debug them and thereby carry out necessary measures to improve their performance.…”

Explainable AI over the Internet of Things (IoT): Overview, State-of-the-Art and Future Directions

“…• Model-based: Model interpretation is the core aspect of any AI-based system. It ensures to provide better decision-making policies with fairness, transparency, and accountability of the results, which may give enough confidence to humans [58]. With the value and accuracy interpreted from the models, it can be easy to debug them and thereby carry out necessary measures to improve their performance.…”

Particle Swarm Optimization-Based Model Abstraction and Explanation Generation for a Recurrent Neural Network

“…Firstly, the state space of the abstraction model is significantly reduced, which can effectively save time and space to visit the original RNN, especially when the dataset is large [11]. Secondly, some practical model analysis and verification techniques can be applied to the abstraction model to explain the original RNN [12]. This undoubtedly meets the inherent requirement of transparency in explainability principles.…”