Unverified50% confidenceFactExact time
Bundle adjustment (BA) exploits the sparse structure of the Hessian matrix via the Schur complement (marginalization) to decompose large-scale problems into smaller subproblems, thereby reducing computational complexity.
1
Sources
50%
Confidence
Long-term
Relevance
8/9/2026
First Seen
Sources
视觉SLAM学习路线图全解析：从数学基础到工程实践
Related Claims
UnverifiedLow-rank decomposition and quantization compression are orthogonally complementary dimensions; quantizing the smaller matrices A and B separately results in smaller errors, and the combined compression ratio can reach several times that of either method alone71% similarUnverifiedK-quant 系列引入了分块量化策略，对权重矩阵中不同的 block 使用不同的缩放参数以减少量化误差69% similarUnverifiedGPTQ（2022）引入逐层Hessian矩阵近似，通过最小化权重量化前后的输出重建误差优化量化参数68% similarUnverified分组量化将权重矩阵按固定大小（通常32或64个元素）分组，每组独立计算量化缩放因子以减少精度损失67% similarUnverified应优先融合访存受限的算子，对算力受限的大型算子（如大矩阵乘法）谨慎融合67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/714092API
curl https://kongchang.com/api/v1/knowledge/claims/714092MCP
get_claim(id=714092)