A novel family of auxiliary tasks based on the successor measure to improve the representations that deep reinforcement learning agents acquire Aneesh Tickoo Artificial Intelligence Category – MarkTechPost
In deep reinforcement learning, an agent uses a neural network to map observations to a policy or return prediction. This network’s function is to turn observations into a sequence of progressively finer characteristics, which the final layer then linearly combines to get the desired… Read More »A novel family of auxiliary tasks based on the successor measure to improve the representations that deep reinforcement learning agents acquire Aneesh Tickoo Artificial Intelligence Category – MarkTechPost