Unverified50% confidenceFactExact time
The online estimation algorithm studied in the paper is a single-pass stochastic gradient algorithm with a constant learning rate γ, in the recursive form θ_t = θ_{t-1} + γ * ψ(Y_t, θ_{t-1}), where ψ is the expectile score function.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
Unverified重参数化技巧将采样表示为 z = μ + σ × ε（ε从标准正态分布采样），使随机性外置于ε，从而让μ和σ的梯度可正常反向传播实现端到端训练67% similarUnverifiedThe researchers proposed a credal-set-based method in which the learner's prediction output is no longer a single probability distribution but a set of possible distributions presented in the form of a credal set.64% similarVerifiedResearch-oriented machine learning requires linear algebra, probability and statistics, and optimization theory as mathematical foundations.63% similarUnverified策略梯度定理证明梯度可表示为∇J(θ) = E[∇log π(a|s;θ) · Q^π(s,a)]，使用log-derivative trick将对概率分布的求导转化为对log概率的求导乘以采样值63% similarUnverifiedThe paper uses random matrix theory to derive deterministic equivalents of the population error for an optimally trained teacher model and a student model trained via gradient flow62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1023451API
curl https://kongchang.com/api/v1/knowledge/claims/1023451MCP
get_claim(id=1023451)