The authors instantiate a generalized-Fano framework using squared Hellinger distance to derive explicit Bayesian CVaR lower bounds for interactive decision problems including Gaussian bandits.
Optimization of conditional value- at-risk
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
SafeSABR cuts severe-stall sessions in Starlink video streaming from 22.8% to 7.2% and worst-5% rebuffering from 54.30 s to 22.68 s at 1.8% QoE cost via behavior-cloning pretraining, risk-calibrated RL, and safe-capacity auditing.
citing papers explorer
-
Instantiating Bayesian CVaR lower bounds in Interactive Decision Making Problems
The authors instantiate a generalized-Fano framework using squared Hellinger distance to derive explicit Bayesian CVaR lower bounds for interactive decision problems including Gaussian bandits.
-
SafeSABR: Risk-Calibrated Adaptive Bitrate Streaming over Starlink Networks
SafeSABR cuts severe-stall sessions in Starlink video streaming from 22.8% to 7.2% and worst-5% rebuffering from 54.30 s to 22.68 s at 1.8% QoE cost via behavior-cloning pretraining, risk-calibrated RL, and safe-capacity auditing.