Unverified50% confidenceFactExact time
双重下降现象最早由Belkin等人在2019年的论文《Reconciling modern machine learning practice and the bias-variance trade-off》中系统描述
1
Sources
50%
Confidence
Long-term
Relevance
8/11/2026
First Seen
Sources
Related Claims
VerifiedGradient descent is the most fundamental optimization algorithm in machine learning; a learning rate that is too large causes oscillation, while one that is too small results in slow convergence.66% similarUnverifiedThe double descent phenomenon occurs not only along the model size dimension, but can also appear with changes in training epochs and data volume.65% similarUnverifiedIn 2019, Belkin et al. published a paper in PNAS formally describing the double descent phenomenon, whereby test error begins to decrease again once the number of model parameters exceeds the interpolation threshold.64% similarVerified推测解码、模型蒸馏、KV Cache优化等推理优化技术的持续进步正在使单位推理成本快速下降64% similarVerifiedThe double descent phenomenon indicates that when the number of model parameters far exceeds the amount of data, the test error may actually decrease again.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/731055API
curl https://kongchang.com/api/v1/knowledge/claims/731055MCP
get_claim(id=731055)