𝘚𝘭𝘰𝘸 𝘣𝘶𝘵 𝘴𝘵𝘦𝘢𝘥𝘺

딥러닝 학계 전반에 혁신적인 돌풍을 몰고 온 논문,Attention is All You Need: https://dl.acm.org/doi/10.5555/3295222.3295349 Attention is all you need | Proceedings of the 31st International Conference on Neural Information Processing SystemsPublication History Published: 04 December 2017dl.acm.org BERT 논문도 컨퍼런스 논문에다 거의 Encoder 구조를 그대로 갔다 썼기 때문에모델의 구조적인 부분은 나와있지 않아 어쩔 수 없이 Transformer 논문도 같이 공부하게 되었고,읽은 김에 정리해서 기록으로 남..

남들 DeepSeek 읽을때 이제서야 BERT 읽고 정리한다https://arxiv.org/abs/1810.04805 BERT: Pre-training of Deep Bidirectional Transformers for Language UnderstandingWe introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers. Unlike recent language representation models, BERT is designed to pre-train deep bidirectional representations from ..