Unverified50% confidenceFactExact time
When d=1, the orthogonal constraint degenerates to spherical optimization, i.e., the direction of each row lies on its own unit sphere, which by itself already outperforms the tuned Muon optimizer.
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGaussian splatting DFT uses the Löwdin orthogonalization technique to project the optimization step back onto the feasible manifold to satisfy the orbital orthonormality constraint.66% similarUnverified在高维参数空间中，鞍点比局部极小值更为普遍，现代优化器通过引入动量机制能有效冲过鞍点区域66% similarUnverified雅可比线性化（在平衡点附近做一阶泰勒展开）只在小偏差范围内有效，载荷偏航角较大时PID、LQR等线性控制器性能会严重退化64% similarUnverifiedCurved structure orthogonal to the probe directly affects the model's generalization ability, robustness, and compositional transfer performance in out-of-distribution (OOD) scenarios.64% similarUnverifiedDuring tuning, one should first verify the effect with a single camera movement direction before gradually adding multi-axis motion.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1129390API
curl https://kongchang.com/api/v1/knowledge/claims/1129390MCP
get_claim(id=1129390)