Multi-source transfer learning incurs an intrinsic adaptation cost that can exceed one, with phase transitions separating regimes where bias-agnostic estimators match oracle performance from those where they cannot.
12 ON CHANGE OF MEASURE INEQUALITIES FOR f -DIVERGENCES Pierre Alquier and Benjamin Guedj
12 Pith papers cite this work, alongside 24 external citations. Polarity classification is still indexing.
12
Pith papers citing it
24
external citations · external index
representative citing papers
AEW achieves the excess risk bound T log(M)/(n+1) in expectation for sufficiently large constant temperature T under i.i.d. random design and bounded L-Lipschitz μ-strongly convex losses.
Proves that Rademacher complexity of depth-d compositional trees over finite operator vocabulary is controlled by (K b L)^{d} / sqrt(n) under Lipschitz conditions on operators.
ALU uses public data to suppress unlearning cost quadratically while characterizing distribution mismatch effects, enabling mass unlearning with maintained utility.
PAC-Bayesian bounds are derived for quadratic closed-loop control via SLS parameterization, yielding Chernoff certificates for posteriors over responses, a mean-response deployment result, and a data-driven learning algorithm.
Prompt-conditioned LLMs face irreducible error floors from language's limited information capacity and alignment constraints, proven via PAC-Bayes bounds on bilevel cheap-talk games for certain task families.
PAC-Bayes bounds for Gibbs posteriors are obtained via singular learning theory, producing explicit and tighter posterior-averaged risk bounds that adapt to data structure in overparameterized models.
A distributionally robust PAC-Bayesian approach derives sub-Gaussian loss proxies and performance bounds tied to closed-loop operator norms via system level synthesis, enabling optimization-based safety certificates for controllers facing sim-to-real gaps.
Federated learning trains private local randomised predictors whose aggregation yields a global predictor with nonvacuous PAC-Bayesian generalisation bounds and near-centralized accuracy.
Derives f-divergence change-of-measure inequalities via Legendre transform to extend PAC-Bayes bounds beyond the classical Donsker-Varadhan setting.
MLP activation sparsity equals augmented flatness divided by input norm times gradient; the ratio falls during training and can be reduced further by three plug-and-play changes, yielding higher sparsity on ImageNet and C4.
A unified DPI-based framework yields novel change-of-measure inequalities that produce tighter high-probability generalization bounds.
citing papers explorer
-
The Statistical Cost of Adaptation in Multi-Source Transfer Learning
Multi-source transfer learning incurs an intrinsic adaptation cost that can exceed one, with phase transitions separating regimes where bias-agnostic estimators match oracle performance from those where they cannot.
-
Aggregation with Exponential Weights is Optimal in Expectation
AEW achieves the excess risk bound T log(M)/(n+1) in expectation for sufficiently large constant temperature T under i.i.d. random design and bounded L-Lipschitz μ-strongly convex losses.
-
Sample Complexity of Scientific Discovery: PAC Learnability of Compositional Function Trees
Proves that Rademacher complexity of depth-d compositional trees over finite operator vocabulary is controlled by (K b L)^{d} / sqrt(n) under Lipschitz conditions on operators.
-
Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data
ALU uses public data to suppress unlearning cost quadratically while characterizing distribution mismatch effects, enabling mass unlearning with maintained utility.
-
PAC-Bayesian Certificates for Quadratic Closed-Loop Control
PAC-Bayesian bounds are derived for quadratic closed-loop control via SLS parameterization, yielding Chernoff certificates for posteriors over responses, a mean-response deployment result, and a data-driven learning algorithm.
-
On the Limits of Prompt-Conditioned Language Models as General-Purpose Learners
Prompt-conditioned LLMs face irreducible error floors from language's limited information capacity and alignment constraints, proven via PAC-Bayes bounds on bilevel cheap-talk games for certain task families.
-
PAC-Bayes Bounds for Gibbs Posteriors via Singular Learning Theory
PAC-Bayes bounds for Gibbs posteriors are obtained via singular learning theory, producing explicit and tighter posterior-averaged risk bounds that adapt to data structure in overparameterized models.
-
Distributionally Robust PAC-Bayesian Control
A distributionally robust PAC-Bayesian approach derives sub-Gaussian loss proxies and performance bounds tied to closed-loop operator norms via system level synthesis, enabling optimization-based safety certificates for controllers facing sim-to-real gaps.
-
Federated Learning with Nonvacuous Generalisation Bounds
Federated learning trains private local randomised predictors whose aggregation yields a global predictor with nonvacuous PAC-Bayesian generalisation bounds and near-centralized accuracy.
-
Change of measure through the Legendre transform
Derives f-divergence change-of-measure inequalities via Legendre transform to extend PAC-Bayes bounds beyond the classical Donsker-Varadhan setting.
-
Towards the Connection between Activation Sparsity and Flat Minima
MLP activation sparsity equals augmented flatness divided by input norm times gradient; the ratio falls during training and can be reduced further by three plug-and-play changes, yielding higher sparsity on ImageNet and C4.
-
Tighter Information-Theoretic Generalization Bounds via a Novel Class of Change of Measure Inequalities
A unified DPI-based framework yields novel change-of-measure inequalities that produce tighter high-probability generalization bounds.