Unverified50% confidenceFactExact time
The universal approximation theorem for neural networks states that a sufficiently wide or sufficiently deep network can approximate any continuous function, but does not answer the question of what network size is required to achieve a specific accuracy.
1
Sources
50%
Confidence
Long-term
Relevance
10/5/2026
First Seen
Sources
Related Entities
Related Claims
Unverified反向传播机制要求整个计算图必须是可微分的，且在极深网络中存在梯度消失/爆炸问题74% similarUnverifiedThe performance of neural network inference is often memory-bound rather than compute-bound.73% similarUnverified仅无约束地最大化SWD会导致神经网络扭曲数据结构得到退化解，因此需引入几何约束在分离性与保真度之间权衡71% similarUnverifiedThe paper proves that there exist networks of constant depth and size Õ_{n,d}(p^{-9}ε^{-max{d/n,2}}log(1/δ)) that satisfy the probabilistic uniform approximation requirement with probability at least 1-δ.71% similarUnverifiedMPC在每个时间步求解有限时域优化问题并显式处理硬约束，这是端到端神经网络难以可靠保证的特性71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/975780API
curl https://kongchang.com/api/v1/knowledge/claims/975780MCP
get_claim(id=975780)