The challenge of developing powerful and general Reinforcement Learning (RL) agents has received increasing attention in recent years. Much of this effort has focused on the single-agent setting, in which an agent maximizes a predefined extrinsic reward function. However, a long-term question inevitably arises: how will such independent agents cooperate when they are continually learning and acting in a shared multi-agent environment? Observing that humans often provide incentives to influence others' behavior, we propose to equip each RL agent in a multi-agent environment with the ability to give rewards directly to other agents, using a learned incentive function. Each agent learns its own incentive function by explicitly accounting for its impact on the learning of recipients and, through them, the impact on its own extrinsic objective. We demonstrate in experiments that such agents significantly outperform standard RL and opponent-shaping agents in challenging general-sum Markov games, often by finding a near-optimal division of labor. Our work points toward more opportunities and challenges along the path to ensure the common good in a multi-agent future.

In this paper we present a Bayesian reinforcement learning framework that allows robotic manipulators to adaptively recover from random mechanical failures autonomously, hence being survivable. To this end, we formulate the framework of Bayesian Policy Morphing (BPM) that enables a robot agent to self-modify its learned policy after the diminution of its maneuvering dimensionality. We build upon existing actor-critic framework, and extend it to perform policy gradient updates as posterior learning, taking past policy updates as prior distributions. We show that policy search, in the direction biased by prior experience, significantly improves learning efficiency in terms of sampling requirements. We demonstrate our results on an 8-DOF robotic arm with our algorithm of BPM, while intentionally disabling random joints with different damage types like unresponsive joints, constant offset errors and angular imprecision. Our results have shown that, even with physical damages, the robotic arm can still successfully maintain its functionality to accurately locate and grasp a given target object.

In this paper, we propose a new algorithm PPG (Proximal Policy Gradient), which is close to both VPG (vanilla policy gradient) and PPO (proximal policy optimization). The PPG objective is a partial variation of the VPG objective and the gradient of the PPG objective is exactly same as the gradient of the VPG objective. To increase the number of policy update iterations, we introduce the advantage-policy plane and design a new clipping strategy. We perform experiments in OpenAI Gym and Bullet robotics environments for ten random seeds. The performance of PPG is comparable to PPO, and the entropy decays slower than PPG. Thus we show that performance similar to PPO can be obtained by using the gradient formula from the original policy gradient theorem.

Learning and planning with latent space dynamics has been shown to be useful for sample efficiency in model-based reinforcement learning (MBRL) for discrete and continuous control tasks. In particular, recent work, for discrete action spaces, demonstrated the effectiveness of latent-space planning via Monte-Carlo Tree Search (MCTS) for bootstrapping MBRL during learning and at test time. However, the potential gains from latent-space tree search have not yet been demonstrated for environments with continuous action spaces. In this work, we propose and explore an MBRL approach for continuous action spaces based on tree-based planning over learned latent dynamics. We show that it is possible to demonstrate the types of bootstrapping benefits as previously shown for discrete spaces. In particular, the approach achieves improved sample efficiency and performance on a majority of challenging continuous-control benchmarks compared to the state-of-the-art.

Imitation Learning (IL) algorithms aim to learn optimal behavior by mimicking expert demonstrations. Perhaps the simplest IL method is Behavioral Cloning (BC) [Pomerleau, 1991] which ignores the dynamics of the underlying Markov Decision Process (MDP) that generated the demonstrations, and treats IL as a supervised learning problem of predicting optimal actions given states. Prior work showed that if the learned policy incurs a small BC loss, the worst case performance gap between the expert and imitator grows quadratically with the number of decision steps [Ross and Bagnell, 2010, Ross et al., 2011a]. The crux of their argument is that policies that are "close" as measured by BC loss can induce disastrously different distributions over states when deployed in the environment. One family of solutions to mitigating such compounding errors is Interactive IL [Ross et al., 2011b, 2013, Guo et al., 2014], which involves running the imitator's policy and collecting corrective actions from an interactive expert. However, interactive expert queries can be expensive and are seldom available. Another family of approaches [Ho and Ermon, 2016, Fu et al., 2017, Ke et al., 2020, Kostrikov et al., 2020, Kim and Park, 2018, Wang et al., 2017] that have gained much traction is to directly minimize a statistical distance between state-action distributions induced by policies of the expert and imitator, i.e the occupancy measures ρ

The value function lies in the heart of Reinforcement Learning (RL), which defines the long-term evaluation of a policy in a given state. In this paper, we propose Policy-extended Value Function Approximator (PeVFA) which extends the conventional value to be not only a function of state but also an explicit policy representation. Such an extension enables PeVFA to preserve values of multiple policies in contrast to a conventional one with limited capacity for only one policy, inducing the new characteristic of \emph{value generalization among policies}. From both the theoretical and empirical lens, we study value generalization along the policy improvement path (called local generalization), from which we derive a new form of Generalized Policy Iteration with PeVFA to improve the conventional learning process. Besides, we propose a framework to learn the representation of an RL policy, studying several different approaches to learn an effective policy representation from policy network parameters and state-action pairs through contrastive learning and action prediction. In our experiments, Proximal Policy Optimization (PPO) with PeVFA significantly outperforms its vanilla counterpart in MuJoCo continuous control tasks, demonstrating the effectiveness of value generalization offered by PeVFA and policy representation learning.

This paper proposes a lexicographic Deep Reinforcement Learning (DeepRL)-based approach to chance-constrained Markov Decision Processes, in which the controller seeks to ensure that the probability of satisfying the constraint is above a given threshold. Standard DeepRL approaches require i) the constraints to be included as additional weighted terms in the cost function, in a multi-objective fashion, and ii) the tuning of the introduced weights during the training phase of the Deep Neural Network (DNN) according to the probability thresholds. The proposed approach, instead, requires to separately train one constraint-free DNN and one DNN associated to each constraint and then, at each time-step, to select which DNN to use depending on the system observed state. The presented solution does not require any hyper-parameter tuning besides the standard DNN ones, even if the probability thresholds changes. A lexicographic version of the well-known DeepRL algorithm DQN is also proposed and validated via simulations.

Groundbreaking successes have been achieved by Deep Reinforcement Learning (DRL) in solving practical decision-making problems. Robotics, in particular, can involve high-cost hardware and human interactions. Hence, scrupulous evaluations of trained models are required to avoid unsafe behaviours in the operational environment. However, designing metrics to measure the safety of a neural network is an open problem, since standard evaluation parameters (e.g., total reward) are not informative enough. In this paper, we present a semi-formal verification approach for decision-making tasks, based on interval analysis, that addresses the computational demanding of previous verification frameworks and design metrics to measure the safety of the models. Our method obtains comparable results over standard benchmarks with respect to formal verifiers, while drastically reducing the computation time. Moreover, our approach allows to efficiently evaluate safety properties for decision-making models in practical applications such as mapless navigation for mobile robots and trajectory generation for manipulators.

Policy optimization in reinforcement learning requires the selection of numerous hyperparameters across different environments. Fixing them incorrectly may negatively impact optimization performance leading notably to insufficient or redundant learning. Insufficient learning (due to convergence to local optima) results in under-performing policies whilst redundant learning wastes time and resources. The effects are further exacerbated when using single policies to solve multi-task learning problems. In this paper, we study how the Evidence Lower Bound (ELBO) used in Variational Auto-Encoders (VAEs) is affected by the diversity of image samples. Different tasks or setups in visual reinforcement learning incur varying diversity. We exploit the ELBO to create an auto-tuning technique in self-supervised reinforcement learning. Our approach can auto-tune three hyperparameters: the replay buffer size, the number of policy gradient updates during each epoch, and the number of exploration steps during each epoch. We use the state-of-the-art self-supervised robotic learning framework (Reinforcement Learning with Imagined Goals (RIG) using Soft Actor-Critic) as baseline for experimental verification. Experiments show that our method can auto-tune online and yields the best performance at a fraction of the time and computational resources. Code, video, and appendix for simulated and real-robot experiments can be found at http://www.JuanRojas.net/autotune.

Model-free reinforcement learning (RL) has been an active area of research and provides a fundamental framework for agent-based learning and decision-making in artificial intelligence. In this paper, we review a specific subset of this literature, namely work that utilizes optimization criteria based on average rewards, in the infinite horizon setting. Average reward RL has the advantage of being the most selective criterion in recurrent (ergodic) Markov decision processes. In comparison to widely-used discounted reward criterion, it also requires no discount factor, which is a critical hyperparameter, and properly aligns the optimization and performance metrics. Motivated by the solo survey by Mahadevan (1996a), we provide an updated review of work in this area and extend it to cover policy-iteration and function approximation methods (in addition to the value-iteration and tabular counterparts). We also identify and discuss opportunities for future work.