Unverified50% confidenceFactExact time
In Bayesian neural networks with continuously increasing width M, the KL divergence regularization term in the ELBO gradually overwhelms the expected log-likelihood term, causing the variational predictive distribution to degenerate into the prior predictive distribution.
1
Sources
50%
Confidence
Long-term
Relevance
10/8/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAs network width M increases, the KL term grows linearly with the number of parameters while the expected log-likelihood term grows relatively more slowly, which is the fundamental mathematical cause of the prior dominance phenomenon.75% similarVerifiedVanishing gradient refers to the exponential decay of gradient values during backpropagation in deep neural networks, causing parameters near the input layer to receive almost no updates.66% similarUnverified深度神经网络天然倾向于过拟合训练分布，对分布外样本的处理能力往往大幅下降65% similarUnverifiedBayesian Knowledge Tracing (BKT) is relatively insensitive to threshold changes, whereas neural network models become noticeably more selective as the threshold increases.64% similarUnverified从KL散度角度看，每一代合成数据与真实分布之间的散度都在单调递增，且误差会在迭代中累积64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/987029API
curl https://kongchang.com/api/v1/knowledge/claims/987029MCP
get_claim(id=987029)