arXiv:2606.25012v1 Announce Type: new Abstract: Many reinforcement learning (RL) problems in the infinite-horizon average-reward setting require optimizing multiple conflicting objectives while satisfying multiple safety constraints. A common approach is concave scalarization, where the agent maximizes a utility $ f(J^\pi_{r_1}, \ldots, J^\pi_{r_M}) $ subject to a scalarized constraint $ g(J^\pi_{c_1}, \ldots, J^\pi_{c_N}) \ge 0 $, where $J^\pi_{r_m}$ and $J^\pi_{c_n}$ denote the average-reward and cost under policy $\pi$. However, the nonlinearity of $f$ and $g$ introduces bias in policy-grad
Source: arXiv cs.LG — read the full report at the original publisher.