AI RESEARCH

Conservative Continuous-Time Treatment Optimization

arXiv CS.LG

ArXi:2603.16789v1 Announce Type: new We develop a conservative continuous-time stochastic control framework for treatment optimization from irregularly sampled patient trajectories. The unknown patient dynamics are modeled as a controlled stochastic differential equation with treatment as a continuous-time control. Naive model-based optimization can exploit model errors and propose out-of- controls, so optimizing the estimated dynamics may not optimize the true dynamics.