Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Ascent Conditions for Length-Normalized Single-Rollout Policy Updates
Levin David Schwab
Dividing each trajectory's score contribution by its realized length can change the direction of a reward-centered policy update. In a finite trajectory model, we derive an exact a…
cs.LG2026
Marginal Log-Likelihood Increments under Dirichlet-Smoothed Markov Estimation
Levin David Schwab
For a Dirichlet-smoothed transition model, the effect of adding one workflow trace to the training archive is an exact change in reference-weighted log likelihood. We derive that c…