Saved in:
Bibliographic Details
Main Authors: Bongole, Raghav, Oechtering, Tobias J., Skoglund, Mikael
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.12027
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912830001774592
author Bongole, Raghav
Oechtering, Tobias J.
Skoglund, Mikael
author_facet Bongole, Raghav
Oechtering, Tobias J.
Skoglund, Mikael
contents Interactive statistical decision making (ISDM) features algorithm-dependent data generated through interaction. Existing information-theoretic lower bounds in ISDM largely target expected risk, while tail-sensitive objectives are less developed. We generalize the interactive Fano framework of Chen et al. by replacing the hard success event with a randomized one-bit statistic representing an arbitrary bounded transform of the loss. This yields a Bernoulli f-divergence inequality, which we invert to obtain a two-sided interval for the transform, recovering the previous result as a special case. Instantiating the transform with a bounded hinge and using the Rockafellar-Uryasev representation, we derive lower bounds on the prior-predictive (Bayesian) CVaR of bounded losses. For KL divergence with the mixture reference distribution, the bound becomes explicit in terms of mutual information via Pinsker's inequality.
format Preprint
id arxiv_https___arxiv_org_abs_2601_12027
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Generalizing the Fano inequality further
Bongole, Raghav
Oechtering, Tobias J.
Skoglund, Mikael
Information Theory
Interactive statistical decision making (ISDM) features algorithm-dependent data generated through interaction. Existing information-theoretic lower bounds in ISDM largely target expected risk, while tail-sensitive objectives are less developed. We generalize the interactive Fano framework of Chen et al. by replacing the hard success event with a randomized one-bit statistic representing an arbitrary bounded transform of the loss. This yields a Bernoulli f-divergence inequality, which we invert to obtain a two-sided interval for the transform, recovering the previous result as a special case. Instantiating the transform with a bounded hinge and using the Rockafellar-Uryasev representation, we derive lower bounds on the prior-predictive (Bayesian) CVaR of bounded losses. For KL divergence with the mixture reference distribution, the bound becomes explicit in terms of mutual information via Pinsker's inequality.
title Generalizing the Fano inequality further
topic Information Theory
url https://arxiv.org/abs/2601.12027