PS-PPO samples a per-trajectory cutoff and importance-weights truncated gradients, preserving the full critic-free update in expectation while cutting RLHF training compute and memory.
The PAV algorithm optimizes binary proper scoring rules
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
There has been much recent interest in application of the pool-adjacent-violators (PAV) algorithm for the purpose of calibrating the probabilistic outputs of automatic pattern recognition and machine learning algorithms. Special cost functions, known as proper scoring rules form natural objective functions to judge the goodness of such calibration. We show that for binary pattern classifiers, the non-parametric optimization of calibration, subject to a monotonicity constraint, can be solved by PAV and that this solution is optimal for all regular binary proper scoring rules. This extends previous results which were limited to convex binary proper scoring rules. We further show that this result holds not only for calibration of probabilities, but also for calibration of log-likelihood-ratios, in which case optimality holds independently of the prior probabilities of the pattern classes.
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
PS-PPO: Prefix-Sampling PPO for Critic-Free RLHF
PS-PPO samples a per-trajectory cutoff and importance-weights truncated gradients, preserving the full critic-free update in expectation while cutting RLHF training compute and memory.