Unverified50% confidenceFactExact time
Under correlated inputs, forcing weight sharing may introduce bias, causing the convolutional student to fail to fully recover the teacher parameters even with infinite data, whereas the locally connected network, by not sharing weights, can adapt more flexibly to the data.
1
Sources
50%
Confidence
Long-term
Relevance
10/22/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSpatial correlation reduces the advantage of weight sharing: the asymptotic error of the convolutional student rises as the input spatial correlation c increases, whereas the asymptotic error of the locally connected student is independent of c and is overall lower at the same learning rate.74% similarUnverifiedCommunication primitives in distributed training (such as AllReduce gradient aggregation) can introduce non-deterministic accumulated errors due to differences in parallelism configuration and network topology.70% similarVerifiedDuring cross-domain fine-tuning, an excessively large learning rate will destroy pre-trained weights, while an excessively small one will result in almost no learning, requiring a careful trade-off.66% similarUnverified当目标数据域与源数据域差异悬殊时，预训练权重可能产生负迁移（Negative Transfer），反而干扰目标域的学习66% similarUnverified该神经网络能力上限被Minimax数据锁定，最多只能逼近教师算法水平而无法超越64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1083915API
curl https://kongchang.com/api/v1/knowledge/claims/1083915MCP
get_claim(id=1083915)