TY - GEN
T1 - Application of Neuro-Symbolic Reasoning in Natural Language Processing
AU - Aithal, Shivani G.
AU - Rao, Abishek B.
AU - Chandrakala, C. B.
AU - Singh, Sanjay
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Commonsense reasoning is a crucial part of human intelligence. It has been a goal in the study of artificial intelligence to develop machines mimicking commonsense reasoning at par with humans. The deep learning models have advanced in many natural language processing tasks. However, they lack logical reasoning ability. Symbolic artificial intelligence has the potential reasoning ability wherein logic is used to reason efficiently. This led to the rise of neuro-symbolic reasoning, which is an integration of symbolic logic with deep learning. In this paper, we have experimented with the BERT, RoBERTa and GPT-J language models for question answering task to check their reasoning ability and found that they lack the reasoning ability. This paper compares the performance of the BERT, RoBERTa and GPT-J question-answering models with the neuro-symbolic reasoning model using logical neural networks. It is found that the neuro-symbolic reasoning model has the reasoning ability like humans, which infers that the neuro-symbolic reasoning model can effectively handle logical reasoning and is a key to the future of artificial intelligence.
AB - Commonsense reasoning is a crucial part of human intelligence. It has been a goal in the study of artificial intelligence to develop machines mimicking commonsense reasoning at par with humans. The deep learning models have advanced in many natural language processing tasks. However, they lack logical reasoning ability. Symbolic artificial intelligence has the potential reasoning ability wherein logic is used to reason efficiently. This led to the rise of neuro-symbolic reasoning, which is an integration of symbolic logic with deep learning. In this paper, we have experimented with the BERT, RoBERTa and GPT-J language models for question answering task to check their reasoning ability and found that they lack the reasoning ability. This paper compares the performance of the BERT, RoBERTa and GPT-J question-answering models with the neuro-symbolic reasoning model using logical neural networks. It is found that the neuro-symbolic reasoning model has the reasoning ability like humans, which infers that the neuro-symbolic reasoning model can effectively handle logical reasoning and is a key to the future of artificial intelligence.
UR - http://www.scopus.com/inward/record.url?scp=85146725682&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85146725682&partnerID=8YFLogxK
U2 - 10.1109/CICT56698.2022.9997814
DO - 10.1109/CICT56698.2022.9997814
M3 - Conference contribution
AN - SCOPUS:85146725682
T3 - 2022 IEEE 6th Conference on Information and Communication Technology, CICT 2022
BT - 2022 IEEE 6th Conference on Information and Communication Technology, CICT 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 6th IEEE Conference on Information and Communication Technology, CICT 2022
Y2 - 18 November 2022 through 20 November 2022
ER -