We theoretically establish that our algorithms almost surely converge to locally optimal policies of our safe optimization framework.

We name this problem Safe-RL-SW . Our step-wise violation constraint differs from prior expected violation constraint (Wachi & Sui, 2020; Efroni et al., 2020b; Kalagarla et al., 2021) in two aspects: (i) Minimizing the step-wise violation enables the agent to learn an optimal policy that avoids unsafe regions deterministically,

We name this problem Safe-RL-SW . Our step-wise violation constraint differs from prior expected violation constraint (Wachi & Sui, 2020; Efroni et al., 2020b; Kalagarla et al., 2021) in two aspects: (i) Minimizing the step-wise violation enables the agent to learn an optimal policy that avoids unsafe regions deterministically,

In particular, the non-linear model is often optimized, whereas the reference KF model is not. We argue that both should be optimized similarly, and to that end present the Optimized KF ( OKF).

Neural Information Processing Systems http://nips.cc/

To combat this, there has been a surge of interest in algorithmic fairness metrics (Mehrabi et al., 2021).