Multi-task Linear Regression without Eigenvalue Lower Bounds: Adaptivity, Robustness, and Safety

ORID D5Ijcnz1L9 · tags icml2026-repro paper-D5Ijcnz1L9

#StatusPageArtifactClaim excerpt
1VERIFIED 2/201-establishes-in-sample-mse-bound-taskartifactTheorem 2 establishes an in-sample MSE bound for each task j that guarantees saf…
2VERIFIED 2/202-transfer-guarantee-inlier-tasks-minartifactTheorem 2 also shows a transfer guarantee for inlier tasks when B ≲ min(1/ε, m):…
3VERIFIED 2/203-assumption-balancedness-replaces-classical-lowerartifactAssumption 1 (Balancedness) replaces the classical Lower Boundedness of Second M…
4VERIFIED 2/204-extends-in-sample-mse-guarantees-populationartifactTheorem 3 extends the in-sample MSE guarantees of Theorem 2 to population risk v…
5VERIFIED 2/205-extends-same-adaptive-safety-transferartifactTheorem 4 extends the same adaptive safety/transfer MSE guarantees to generalize…
6VERIFIED 2/206-algorithm-solves-joint-convex-objectiveartifactAlgorithm 1 solves a joint convex objective ℒ(Θ)=Σⱼ wⱼ(fⱼ(θⱼ)+λⱼ‖θⱼ-β‖_{Σⱼ}) tha…

Open logbook index · logbook.json