Unverified50% confidenceDecision RuleExact time
The square root scaling formula for the learning rate is: learning rate multiplied by sqrt(M/N); when scaling from 10 environments to 1000, the learning rate is adjusted to 0.0003 * sqrt(100) = 0.003, but this may be too aggressive in ML-Agents, so it is recommended to start by trying 1.5-2 times.
1
Sources
50%
Confidence
Long-term
Relevance
10/13/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAdaGrad于2011年首次引入自适应学习率，对历史梯度平方累积归一化，但其分母单调递增会导致学习率趋近于零73% similarUnverified根据学习金字塔理论（Edgar Dale），被动学习的知识留存率在10%-30%之间，而主动学习的留存率可达50%-90%65% similarUnverifiedThe paper proves that the learned generative distribution satisfies a finite-sample error bound in Wasserstein-p distance: W_p ≲ n^(-1/d) + n^(-1/(2p))(log(1/ξ))^(1/(2p)).65% similarUnverified布鲁姆的掌握学习法要求学生达到约80%的掌握标准才能进入下一单元，取得约1.0 SD的效应量64% similarUnverifiedThe computational complexity of the standard self-attention mechanism is proportional to the square of the context length (O(n²)), meaning extending the context from 100K to 1M theoretically increases computation by 100 times.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1015942API
curl https://kongchang.com/api/v1/knowledge/claims/1015942MCP
get_claim(id=1015942)