Replace pointwise high-order PINN residuals with a stochastic one-step residual evaluated on Brownian transitions. A single scalar network produces the value, gradient, and Hessian by automatic differentiation, and the quadratic centered increment supplies a stochastic probe of the Hessian. Add a terminal gradient penalty so the learned full jet is constrained at the terminal boundary, not only the scalar value.
Train a neural value or latent-dynamics model with temporal-difference targets before enforcing a stiff differential-equation residual, and ramp the physics weight only after the critic has become predictive. For a stochastic dynamical model, the residual is computed using the infinitesimal generator, while terminal, safe, and failure boundary conditions are imposed through separate penalties.
Use a learned quasipotential barrier as feedback for optimizer noise and restart control. Increase stochasticity when training is trapped in a high-loss metastable basin and reduce it near a desirable basin, with switching thresholds determined by the estimated barrier rather than by a fixed patience schedule.
Use the recent history of generator outputs as a controllable training window instead of fixing the replay-memory depth globally. Estimate how quickly each fitness level improves as more same-level examples enter the window, and increase memory only when the measured escape probability improves enough to justify the extra stale data.
Attach uncertainty to neural value targets by estimating the empirical one-step Bellman perturbation and propagating it through the discounted closed-loop transition operator. Use the resulting uncertainty to downweight high-variance Bellman targets or regularize the critic toward conservative predictions, especially in offline or model-based reinforcement learning.
Replace the naive sample variance of correlated rollout returns with a recursive variance target attached to every state-action node or latent rollout node. The target separates uncertainty caused by immediate reward noise, stochastic next-state selection, and uncertainty already present in child value estimates, enabling calibrated heteroscedastic Bellman updates and uncertainty-aware rollout allocation.
Use the paper's finite-dimensional second-moment equations to compute the stationary covariance induced by a Markov-switched recurrent layer before training, then whiten or scale each mode's hidden state using that covariance. This can prevent mode-specific saturation and eliminate a long burn-in period in long-context RNNs and state-space models.