Unverified50% confidenceFactExact time
LIPS-PC parameterizes the measurement operator as a residual correction to the 2D-DCT basis matrix, i.e., D = D_base + Δ, where Δ is trained end-to-end jointly with a compact classifier via a cross-entropy loss.
1
Sources
50%
Confidence
Long-term
Relevance
11/4/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn ridge regression, storing only the predictor costs dp, whereas an exact learning state requires dp + d(d+1)/2, with the additional d(d+1)/2 corresponding to the second-order covariance structure information of the design matrix.66% similarUnverifiedCLIP uses softmax cross-entropy loss with a computational complexity of O(n²), while SigLIP's sigmoid loss allows each image-text pair to be computed independently, eliminating the hard dependency on large-batch training.66% similarUnverifiedThe LIPS-PC method has limitations: at high budgets its advantage weakens or even reverses, possibly because the learned patterns overfit to low-measurement conditions; and the experiments involve only linear measurements and simple classifiers.66% similarUnverifiedDifferential Privacy (DP) clips training gradients and injects calibrated Gaussian noise, allowing the influence of any single training data point's presence or absence on model output to be mathematically bounded.65% similarUnverifiedFor a d×d weight matrix, LoRA decomposes it into two low-rank matrices A (d×r) and B (r×d), reducing trainable parameters from d² to 2dr, where the rank r is typically 8, 16 or 64.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1134327API
curl https://kongchang.com/api/v1/knowledge/claims/1134327MCP
get_claim(id=1134327)