Unverified50% confidenceFactExact time
OU过程的平稳分布为高斯分布，其转移核为高斯分布，使任意时刻的条件分布可以精确计算
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
UnverifiedOU过程具有均值回归性质，其数学形式为 dXt = -αXt dt + σ dWt，通过适当设置参数任意初始分布会在有限时间内指数级收敛到标准高斯分布78% similarUnverifiedThe posterior distribution is obtained via Bayes' theorem: p(θ|x) ∝ p(x|θ)·p(θ), combining prior knowledge with the data likelihood.68% similarUnverified扩散模型生成过程在浮点数连续空间中运算，只在最终步骤才量化为整数，因此归纳偏置天然倾向于平滑渐变输出而非硬边界62% similarUnverifiedLLM的每次调用可以被建模为伯努利试验，N次独立调用的通过次数服从二项分布B(N, p)61% similarVerifiedStable Diffusion is built on the Latent Diffusion Model (LDM) architecture61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/524993API
curl https://kongchang.com/api/v1/knowledge/claims/524993MCP
get_claim(id=524993)