Downloads: 0 | Views: 154
Informative Article | Architecture & Planning | India | Volume 11 Issue 4, April 2022 | Popularity: 4.3 / 10
A Comprehensive Survey of Deep Learning Architectures for Natural Language Processing
Akshata Upadhye
Abstract: In recent years, the advancement of deep learning has led a profound transformation in natural language processing (NLP), leading to significant advancements across a variety of language-related tasks. This survey paper aims to provide an exhaustive examination of deep learning architectures that have been important in NLP tasks. Our survey encompasses of a comprehensive exploration of recurrent neural networks (RNNs), convolutional neural networks (CNNs), and transformer- based models, including notable examples such as BERT and GPT, alongside their various variants. We also provide insights about the evolution of deep learning within the domain of NLP, highlighting key milestones and breakthroughs. Additionally, we discuss the important role of benchmark datasets in facilitating thorough evaluation and benchmarking of NLP models. Finally, we also discuss the some of the challenges impeding the seamless training and deployment of large-scale language models, illuminating issues ranging from data inefficiency to ethical considerations. Through this comprehensive survey, we aim to provide a comprehensive overview of deep learning in NLP, by discussing its evolution, benchmarks, and the challenges that practitioners face during training and deployment.
Keywords: Natural Language Processing, Deep Learning, Neural Networks, Transformer, Ethical Considerations, Model Interpretability
Edition: Volume 11 Issue 4, April 2022
Pages: 1384 - 1388
Make Sure to Disable the Pop-Up Blocker of Web Browser