VIEWS40
Kyle Kastner@kastnerkyle
Additionally "neural thickets" give some interesting empirical evidence that this search for behavior is often nearby in weight space https://arxiv.org/abs/2603.12228
19hViews 40
Taco Cohen's mathematical proof equating KL-regularized reinforcement learning to variational inference's ELBO resurfaces in AI discussions
The framework maps REINFORCE directly to score function estimators
1183193.1K
Original post unavailable.
Sentiment
Sentiment building, check back later.
Cluster Engagement
Posts from X
Most Activity
Most Activity