ML
![[논문 리뷰] An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (ViT)](https://img1.daumcdn.net/thumb/R750x0/?scode=mtistory2&fname=https%3A%2F%2Fblog.kakaocdn.net%2Fdna%2Fb0WvzC%2FbtsOB6LLHno%2FAAAAAAAAAAAAAAAAAAAAAI08pt3u7RqCBMYyh4MQS4MMFO33sZrKKZ82emU4aAO5%2Fimg.png%3Fcredential%3DyqXZFxpELC7KVnFOS48ylbz2pIh7yKj8%26expires%3D1753973999%26allow_ip%3D%26allow_referer%3D%26signature%3Dv9mFHipc2%252FJ4jGxpGMq7ptF4W3U%253D)
[논문 리뷰] An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (ViT)
https://arxiv.org/abs/2010.11929 An Image is Worth 16x16 Words: Transformers for Image Recognition at ScaleWhile the Transformer architecture has become the de-facto standard for natural language processing tasks, its applications to computer vision remain limited. In vision, attention is either applied in conjunction with convolutional networks, or used to reparxiv.orgIntroductionSelf-attention..
![[논문 리뷰] Asynchronous Methods for Deep Reinforcement Learning (A3C)](https://img1.daumcdn.net/thumb/R750x0/?scode=mtistory2&fname=https%3A%2F%2Fblog.kakaocdn.net%2Fdna%2FbQRgkD%2FbtssQatbwNl%2FAAAAAAAAAAAAAAAAAAAAAFkpjBZsKHAyGO87g4-wFPID8_QacfhGtYD6VYWV5wVd%2Fimg.webp%3Fcredential%3DyqXZFxpELC7KVnFOS48ylbz2pIh7yKj8%26expires%3D1753973999%26allow_ip%3D%26allow_referer%3D%26signature%3DCocPfP%252BbScH6aeSOhlNHSifT9ys%253D)
[논문 리뷰] Asynchronous Methods for Deep Reinforcement Learning (A3C)
이번 논문에서는 강화학습을 비동기적이게 학습을 하게 만든 논문을 들고 왔다. 이 논문의 특이점이라고 한다면 보통의 학습에서 쓰이는 GPU를 사용하지 않고 CPU 코어들을 통한 병렬학습을 한다는 것이다. 이를 통해 Atari 벤치마크에서 새로운 기록을 세웠고 다른 도메인에서도 좋은 결과를 보여주는 모습이다. [1602.01783] Asynchronous Methods for Deep Reinforcement Learning (arxiv.org) Asynchronous Methods for Deep Reinforcement Learning We propose a conceptually simple and lightweight framework for deep reinforcement learning that..
부스트캠프 17주차 학습 일지 - Product Serving
사용한 기술 스택들: 1/10 월 학습한 것들: We use cloud service so that the server is up 24 hours. Serverless computing: the server is controlled by the cloud by uploading the code to the cloud Stateless Container: docker image-based server Object Storage: can store many types of files Database: for web/app Data Warehouse: database for data analysis CI: automating build and testing CD: automating deployment Wor..
부스트캠프 8주차 학습 일지 - AI 서비스 개발 기초
사용한 기술 스택들: 11/7 월 학습한 것들: MLOps: ML +Ops(operations) - 업무 자동화 - Machine Learning engineering + data engineering + cloud + infrastructure - 빠른 시간 내에 가장 적은 위험을 부담하며 아이디어 단계부터 production 단계까지 진행할 수 있도록 기술적 마찰 줄이기 Research ML vs Production ML: - static data vs dynamic data - good performance vs fast inference with good performance - SOTA vs stable - offline vs online MLOps components: Model Data, f..

LeNet-5으로 더욱더 정확한 손글씨 분류기 만들기: MNIST-2
사용한 기술 스택들: LimePencil/MNIST: MNIST model trained using various models, implemented in PyTorch (github.com) GitHub - LimePencil/MNIST: MNIST model trained using various models, implemented in PyTorch MNIST model trained using various models, implemented in PyTorch - GitHub - LimePencil/MNIST: MNIST model trained using various models, implemented in PyTorch github.com 오늘은 저번 포스팅에 이어서 손글씨 분류기를 만들어..