Refined Differentially Private Linear Regression via Extension of a Free Lunch Result

2026-04-15T08:51:42Zd4001f0fc7fef23458902b6023cf444e35ee4c86d8077a648e5c4d5cf69ac30e
Gaussian-belief-propagationLDPCPMUTurán-theoryUAVcoding-theorydecentralized-algorithmsdifferential-privacydiffusion-modelsemergency-communicationsfederated-learninglinear-regressionmulti-agent-RLpath-compensationpath-planningpower-systemspreemptive-schedulingsecure-aggregationsecure-gradient-coding','groupwise-keyssemantic-communicationsimplex-transformationstate-estimationsuccessive-interference-cancellationtrapping-setswireless-scheduling

What happened

Collection of new arXiv submissions (Apr 15 2026) across privacy-preserving ML, communications, coding, control, and power-system estimation. Key contributions include: (1) an extension of the “free lunch” simplex transformation for add/remove differential privacy to refine sufficient-statistic estimates for private ordinary least-squares linear regression (and polynomial regression); (2) CAPS-TDPC — a channel-aware preemptive scheduling framework that interrupts truncated diffusion-based semantic generation for early wireless transmission with receiver-side path-compensation; (3) a vectorized

Why it matters

A reviewed impact interpretation has not been published for this record.

Evidence and limitations

Source ID
arxiv_math_it
Record identifier
d4001f0fc7fef23458902b6023cf444e35ee4c86d8077a648e5c4d5cf69ac30e
Enrichment time
2026-04-15T08:51:42Z
AI-assisted enrichment
Yes

This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.