JoonoJoono JoonoJoono

Take HomePriliminariesAbstract7.1 Problem Definitions7.2 Computing Derivatives7.3 Toy Example7.4 Backpropagation Algorithm7.4.1. Backpropagation Algorithm Summary7.4.2. Algorithmic Differentiation7.5. Parameter Initialization7.5.1. He Initialization (ReLU 를 쓴다면 이 initialization 을 쓰세요)7.5.3. Initialization for both forward and backward passTake HomeBackpropagation Algorithm이 어떻게 동작하는지에 대한 이해Forwa..

GitHub - stanford-cs149/asst1: Stanford CS149 -- Assignment 1Stanford CS149 -- Assignment 1. Contribute to stanford-cs149/asst1 development by creating an account on GitHub.https://github.com/stanford-cs149/asst1#program-3-parallel-fractal-generation-using-ispc-20-points문제 설명본 과제는 K-Means clustering algorithm을 사용해서 100만개의 데이터 포인트들을 클러스터링 하는 프로그램을 다룬다. What is K-Means?label이 있는 데이터셋, {xi,yi}i=1N\..

본 포스팅은 Simon J.D. Prince 의 Deep Learning 교재를 스터디하며 정리한 글임을 밝힙니다.https://udlbook.github.io/udlbook/ Take Home모델을 학습하는 것은 파라미터, ϕ\phiϕ 에 대응하는 loss function, L[ϕ]L[\phi]L[ϕ] 를 최소화하는 것으로 생각할 수 있다. Gradient Descent 는 현재 파라미터에서 계산되는 loss의 (해당 지점에서의 uphill) gradient를 계산하고 이의 반대방향인 downhill (gradient에 ×−1\times -1×−1 을 곱하면 됨.) 방향으로 파라미터를 업데이트 한다.non-linear function 에 대한 loss는 non-convex일 확률이 아주아주 높다...

본 포스팅은 Stanford Univ. 의 CS149 2022 fall 수업을 정리한 내용임을 밝힙니다.From smart phones, to multi-core CPUs and GPUs, to the world's largest supercomputers and web sites, parallel processing is ubiquitous in modern computing. The goal of this course is to provide a deep understanding of the fundamental principles and engineering trade-offs involved in designing modern parallel computing systems as well as t..

본 포스팅은 Stanford Univ. 의 CS149 2022 fall 수업을 정리한 내용임을 밝힙니다.From smart phones, to multi-core CPUs and GPUs, to the world's largest supercomputers and web sites, parallel processing is ubiquitous in modern computing. The goal of this course is to provide a deep understanding of the fundamental principles and engineering trade-offs involved in designing modern parallel computing systems as well as t..

본 포스팅은 Stanford Univ. 의 CS149 2022 fall 수업을 정리한 내용임을 밝힙니다.From smart phones, to multi-core CPUs and GPUs, to the world's largest supercomputers and web sites, parallel processing is ubiquitous in modern computing. The goal of this course is to provide a deep understanding of the fundamental principles and engineering trade-offs involved in designing modern parallel computing systems as well as t..

본 포스팅은 Stanford Univ. 의 CS149 2022 fall 수업을 정리한 내용임을 밝힙니다.From smart phones, to multi-core CPUs and GPUs, to the world's largest supercomputers and web sites, parallel processing is ubiquitous in modern computing. The goal of this course is to provide a deep understanding of the fundamental principles and engineering trade-offs involved in designing modern parallel computing systems as well as t..

본 포스팅은 Stanford Univ. 의 CS149 2022 fall 수업을 정리한 내용임을 밝힙니다.From smart phones, to multi-core CPUs and GPUs, to the world's largest supercomputers and web sites, parallel processing is ubiquitous in modern computing. The goal of this course is to provide a deep understanding of the fundamental principles and engineering trade-offs involved in designing modern parallel computing systems as well as t..