No abstract

In this work, we combine Curriculum Learning with Deep Reinforcement Learning to learn without any prior domain knowledge, an end-to-end competitive driving policy for the CARLA autonomous driving simulator. To our knowledge, we are the first to provide consistent results of our driving policy on all towns available in CARLA. Our approach divides the reinforcement learning phase into multiple stages of increasing difficulty, such that our agent is guided towards learning an increasingly better driving policy. The agent architecture comprises various neural networks that complements the main convolutional backbone, represented by a ShuffleNet V2. Further contributions are given by (i) the proposal of a novel value decomposition scheme for learning the value function in a stable way and (ii) an ad-hoc function for normalizing the growth in size of the gradients. We show both quantitative and qualitative results of the learned driving policy.

Autonomous Vehicles promise to transport people in a safer, accessible, and even efficient way. Nowadays, real-world autonomous vehicles are build by large teams from big companies with a tremendous amount of engineering effort. Deep Reinforcement Learning can be used instead, without domain experts, to learn end-to-end driving policies. Here, we combine Curriculum Learning with deep reinforcement learning, in order to learn without any prior domain knowledge, an end-to-end competitive driving policy for the CARLA autonomous driving simulator. To our knowledge, this is the first work which provides consistent results of our driving policy on all the town scenarios provided by CARLA. Moreover, we point out two important issues in reinforcement learning: the former is about learning the value function in a stable way, whereas the latter is related to normalizing the learned advantage function. A proposal of a solution to these problems is provided.

Signal-background classification is a central problem in high-energy physics, that plays a major role for the discovery of new fundamental particles. A recent method—the parametric neural network (pNN)—leverages multiple signal mass hypotheses as an additional input feature to effectively replace a whole set of individual classifiers, each providing (in principle) the best response for the corresponding mass hypothesis. In this work we aim at deepening the understanding of pNNs in light of real-world usage. We discovered several peculiarities of parametric networks, providing intuition, metrics, and guidelines to them. We further propose an alternative parametrization scheme, resulting in a new parametrized neural network architecture: the AffinePNN; along with many other generally applicable improvements, like the balanced training procedure. Finally, we extensively and empirically evaluate our models on the HEPMASS dataset, along its imbalanced version (called HEPMASS-IMB) we provide here for the first time, to further validate our approach. Provided results are in terms of the impact of the proposed design decisions, classification performance, and interpolation capability, as well.

Reinforcement Learning (RL) has already achieved several breakthroughs on complex, highdimensional, and even multi-agent tasks, gaining increasingly interest from not only the research community. Although very powerful in principle, its applicability is still limited to solving games and control problems, leaving plenty opportunities to apply and develop RL algorithms for (but not limited to) scientific domains like physics, and biology. Apart from the domain of interest, the applicability of RL is also limited by numerous difficulties encountered while training agents, like training instabilities and sensitivity to hyperparameters. For such reasons, we propose a modern, modular, simple and understandable Python RL library called reinforce-lib. Our main aim is to enable newcomers, practitioners, and researchers to easily employ RL to solve new scientific problems. Our library is available at https://github.com/Luca96/reinforce-lib.The paper is organized as follows: in section 1 we introduce and motivate our contribution, in sections 2 and 3 we provide a short introduction to the reinforcement learning paradigm, as well as describing three popular families of algorithms, then, in the following section 4 we present our library, finally section 5 concludes our discussion.