… However, the low sample efficiency and difficulty of designing reward functions for DRL …
incorporate human prior knowledge in DRL, in order to improve the sample efficiency and save …

… An exploration strategy is derived prior to training that … scenario involving multiple traffic
participants. The resulting … on parametric representation of the Q function such as deep neural net…

… online learning method to mimic human actions. We design two challenging autonomous
driving … from RL’s behavior policy due to prior knowledge and reasoning ability. Thus, a more …

… RL from scratch fails to learn a policy at all. We also compare against prior works that
incorporate prior data for RL, and show that PARROT substantially outperforms these prior works. …

… to learn a parameterized feature (reward) function. Simulated results demonstrate the desired
driving behaviors of an autonomous vehicle … to use if the prior knowledge of the reward …

… A parameterized model M with parameter ψ is established to predict the transition dynamics
… Lv, “Efficient deep reinforcement learning with imitative expert priors for autonomous driving,…

… no prior works have used this branch of techniques to formulate and solve autonomous driving
… Now considering we are using a parametric function as the policy πφ, for example a deep …

… References [4] addresses the deep learning techniques for AD with … Compared with the task
and prior knowledge, the ego vehicle … Sun, “Parameterized batch reinforcement learning for …

… policy πϕ parameterized by ϕ, … reinforcement learning algorithm with expert demonstrations
is put forward to leverage human prior knowledge, in order to improve the sample efficiency …

… reinforcement learning framework can help address this difficulty. We encode prior knowledge
on the latent states of other drivers … including the un-parameterized belief tracker [24] and …