The paper introduces a capped squared loss for contextual distribution learning, but the central proof relies on an invalid convexity assumption and a flawed Markov step.
Semi-bandit learning for monotone stochastic optimization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Contextual Learning for Stochastic Optimization
The paper introduces a capped squared loss for contextual distribution learning, but the central proof relies on an invalid convexity assumption and a flawed Markov step.