Notice
Recent Posts
Recent Comments
Link
반응형
| 일 | 월 | 화 | 수 | 목 | 금 | 토 |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | ||||
| 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| 11 | 12 | 13 | 14 | 15 | 16 | 17 |
| 18 | 19 | 20 | 21 | 22 | 23 | 24 |
| 25 | 26 | 27 | 28 | 29 | 30 | 31 |
Tags
- 로드
- 업무강도
- how-to
- 일상
- 다중
- serviceworkerversion
- optimizations
- sftp
- 회사
- NN
- 구매리스트
- multiclass-label
- vsCode
- deep-learning
- cross-entropy
- flutter
- 입문
- miniconda
- 변속
- jupyter-notebook
- 안장
- stone-weierstrass
- loadentrypoint
- universal-approximation
- neural-networks
Archives
- Today
- Total
목록universal-approximation (1)
Happy Sisyphe
Why Neural Networks Can Approximate Any Function
Universal Approximation TheoremTheoremLet $f: \mathbb{R}^n \to \mathbb{R}$ be a continuous function defined on a compact domain $D \subseteq \mathbb{R}^n$. For any $\epsilon > 0$, there exists a feedforward neural network with a single hidden layer, using a nonlinear, continuous activation function $\phi: \mathbb{R} \to \mathbb{R}$, such that:$$\sup_{x \in D} |f(x) - \hat{f}(x)| $$where $\hat{f}..
Programming/ML&DL
2024. 12. 17. 15:29