Unverified50% confidenceFactExact time
Zinkevich et al. proved in 2007 that after T iterations, the distance between the average strategy produced by CFR and the Nash equilibrium converges at a rate of O(1/√T).
1
Sources
50%
Confidence
Long-term
Relevance
8/5/2026
First Seen
Sources
自博弈AI攻克多米诺骨牌：MCTS与CFR的抽象化困境
Related Claims
UnverifiedThe probability that at least one instance triggers a bug after N instances run for time T is 1-(1-p)^(N×T), which approaches certainty when N is large enough, even if p is extremely small.61% similarUnverified理论分析表明，在时间步长为T的轨迹上，行为克隆的累积误差以O(T²)的速度增长61% similarUnverifiedWhen d is much greater than n, the minimum-norm solution can be selected via the Moore-Penrose pseudoinverse X^T(XX^T)⁻¹y, and the stability of the solution is improved.60% similarUnverified完整雅可比矩阵的计算代价是O(m×n)次前向传播，实际应用中研究者通常采用随机化SVD或Lanczos迭代等算法近似计算其低秩分解58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/687001API
curl https://kongchang.com/api/v1/knowledge/claims/687001MCP
get_claim(id=687001)