Integrating Bayesian Reasoning and Deep Learning


Tan, Sin Yin and Tan, Shing Chiang and Tee, Connie (2022) Integrating Bayesian Reasoning and Deep Learning. In: Postgraduate Colloquium December 2022, 1-15 December 2022, Multimedia University, Malaysia. (Unpublished)

[img] Text
36_Tan Sin Yin_FIST.pdf - Submitted Version
Restricted to Registered users only

Download (818kB)


Deep learning (DL) has performed remarkable achievements on perception tasks by improving the complex structure with multiple layers. However, its inner structure is well-known as a black box because the users do not know the reasons why a certain decision is made. To overcome this shortcoming, Bayesian reasoning (BR) is proposed to integrate with DL where BR is an expressive knowledge representation, and it has explanation capability to reflect a cause-and-effect relationship. It relies on the probabilities which representing by a powerful probabilistic graphical model, Bayesian Network (BN). The study involved extracting the rule sets from Deep Neural Network (DNN) and extracted rules is used to construct an efficient BN. This proposed framework will aid the users to understand why a solution or decision is recommended based on the explanation provided.

Item Type: Conference or Workshop Item (Poster)
Uncontrolled Keywords: Deep learning, Machine learning
Subjects: Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions: Faculty of Information Science and Technology (FIST)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 16 Dec 2022 07:38
Last Modified: 16 Dec 2022 07:38


Downloads per month over past year

View ItemEdit (login required)