In the last decade, there have been significant advances in multi-agent reinforcement learning (MARL) but there are still numerous challenges, such as high sample complexity and slow convergence to stable policies, that need to be overcome before wide-spread deployment is possible. However, many real-world environments already, in practice, deploy sub-optimal or heuristic approaches for generating policies. An interesting question which arises is how to best use such approaches as advisors to help improve reinforcement learning in multi-agent domains. In this paper, we provide a principled framework for incorporating action recommendations from online sub-optimal advisors in multi-agent settings. We describe the problem of ADvising Multiple Intelligent Reinforcement Agents (ADMIRAL) in nonrestrictive general-sum stochastic game environments and present two novel Q-learning based algorithms: ADMIRAL - Decision Making (ADMIRAL-DM) and ADMIRAL - Advisor Evaluation (ADMIRAL-AE), which allow us to improve learning by appropriately incorporating advice from an advisor (ADMIRAL-DM), and evaluate the effectiveness of an advisor (ADMIRAL-AE). We analyze the algorithms theoretically and provide fixed-point guarantees regarding their learning in general-sum stochastic games. Furthermore, extensive experiments illustrate that these algorithms: can be used in a variety of environments, have performances that compare favourably to other related baselines, can scale to large state-action spaces, and are robust to poor advice from advisors.

Emergency vehicles (EMVs) play a critical role in a city's response to time-critical events such as medical emergencies and fire outbreaks. The existing approaches to reduce EMV travel time employ route optimization and traffic signal pre-emption without accounting for the coupling between route these two subproblems. As a result, the planned route often becomes suboptimal. In addition, these approaches also do not focus on minimizing disruption to the overall traffic flow. To address these issues, we introduce EMVLight in this paper. This is a decentralized reinforcement learning (RL) framework for simultaneous dynamic routing and traffic signal control. EMVLight extends Dijkstra's algorithm to efficiently update the optimal route for an EMV in real-time as it travels through the traffic network. Consequently, the decentralized RL agents learn network-level cooperative traffic signal phase strategies that reduce EMV travel time and the average travel time of non-EMVs in the network. We have carried out comprehensive experiments with synthetic and real-world maps to demonstrate this benefit. Our results show that EMVLight outperforms benchmark transportation engineering techniques as well as existing RL-based traffic signal control methods.

Artificial intelligence has proven that complicated board and video games are no longer the exclusive domain of the human mind. From chess to Go to StarCraft, AI systems that use reinforcement learning algorithms have outperformed human world champions in recent years. But despite the high individual performance of RL agents, they can become frustrating teammates when paired with human players, according to a study by AI researchers at MIT Lincoln Laboratory. The study, which involved cooperation between humans and AI agents in the card game Hanabi, shows that players prefer the classic and predictable rule-based AI systems over complex RL systems. The findings, presented in a paper published on arXiv, highlight some of the underexplored challenges of applying reinforcement learning to real-world situations and can have important implications for the future development of AI systems that are meant to cooperate with humans. Deep reinforcement learning, the algorithm used by state-of-the-art game-playing bots, starts by providing an agent with a set of possible actions in the game, a mechanism to receive feedback from the environment, and a goal to pursue.

Model predictive control (MPC) is increasingly being considered for control of fast systems and embedded applications. However, the MPC has some significant challenges for such systems. Its high computational complexity results in high power consumption from the control algorithm, which could account for a significant share of the energy resources in battery-powered embedded systems. The MPC parameters must be tuned, which is largely a trial-and-error process that affects the control performance, the robustness and the computational complexity of the controller to a high degree. In this paper, we propose a novel framework in which any parameter of the control algorithm can be jointly tuned using reinforcement learning(RL), with the goal of simultaneously optimizing the control performance and the power usage of the control algorithm. We propose the novel idea of optimizing the meta-parameters of MPCwith RL, i.e. parameters affecting the structure of the MPCproblem as opposed to the solution to a given problem. Our control algorithm is based on an event-triggered MPC where we learn when the MPC should be re-computed, and a dual mode MPC and linear state feedback control law applied in between MPC computations. We formulate a novel mixture-distribution policy and show that with joint optimization we achieve improvements that do not present themselves when optimizing the same parameters in isolation. We demonstrate our framework on the inverted pendulum control task, reducing the total computation time of the control system by 36% while also improving the control performance by 18.4% over the best-performing MPC baseline.

Reinforcement Learning (RL) agents can learn to solve complex sequential decision making tasks by interacting with the environment. However, sample efficiency remains a major challenge. In the field of multi-goal RL, where agents are required to reach multiple goals to solve complex tasks, improving sample efficiency can be especially challenging. On the other hand, humans or other biological agents learn such tasks in a much more strategic way, following a curriculum where tasks are sampled with increasing difficulty level in order to make gradual and efficient learning progress. In this work, we propose a method for automatic goal generation using a dynamical distance function (DDF) in a self-supervised fashion. DDF is a function which predicts the dynamical distance between any two states within a markov decision process (MDP). With this, we generate a curriculum of goals at the appropriate difficulty level to facilitate efficient learning throughout the training process. We evaluate this approach on several goal-conditioned robotic manipulation and navigation tasks, and show improvements in sample efficiency over a baseline method which only uses random goal sampling.

Processing raw sensory inputs is crucial for applying deep RL algorithms to real-world problems. For example, autonomous vehicles must make decisions about how to drive safely given information flowing from cameras, radar, and microphones about the conditions of the road, traffic signals, and other cars and pedestrians. However, direct "end-to-end" RL that maps sensor data to actions (Figure 1, left) can be very difficult because the inputs are high-dimensional, noisy, and contain redundant information. Instead, the challenge is often broken down into two problems (Figure 1, right): (1) extract a representation of the sensory inputs that retains only the relevant information, and (2) perform RL with these representations of the inputs as the system state. A wide variety of algorithms have been proposed to learn lossy state representations in an unsupervised fashion (see this recent tutorial for an overview).

The recent advancements in Artificial Intelligence (AI) have been a boon for various applied fields. Artificially intelligent systems today are present everywhere around us, chatbots, for instance, replying back to queries in accordance with the interaction made, processing a result in return. Recently, one of the most popular areas of research in artificial intelligence has been in the field of video games. Challenging yet easy to formalize, this platform can be very well used to develop new AI methods and measure how well they work. Video games can also help demonstrate that machines today are capable of behaviour that is thought to require intelligence without putting human lives or property at risk.

We study risk-sensitive reinforcement learning (RL) based on the entropic risk measure. Although existing works have established non-asymptotic regret guarantees for this problem, they leave open an exponential gap between the upper and lower bounds. We identify the deficiencies in existing algorithms and their analysis that result in such a gap. To remedy these deficiencies, we investigate a simple transformation of the risk-sensitive Bellman equations, which we call the exponential Bellman equation. The exponential Bellman equation inspires us to develop a novel analysis of Bellman backup procedures in risk-sensitive RL algorithms, and further motivates the design of a novel exploration mechanism. We show that these analytic and algorithmic innovations together lead to improved regret upper bounds over existing ones.

Deep reinforcement learning (RL) has made it possible to solve complex robotics problems using neural networks as function approximators. However, the policies trained on stationary environments suffer in terms of generalization when transferred from one environment to another. In this work, we use Robust Markov Decision Processes (RMDP) to train the drone control policy, which combines ideas from Robust Control and RL. It opts for pessimistic optimization to handle potential gaps between policy transfer from one environment to another. The trained control policy is tested on the task of quadcopter positional control. RL agents were trained in a MuJoCo simulator. During testing, different environment parameters (unseen during the training) were used to validate the robustness of the trained policy for transfer from one environment to another. The robust policy outperformed the standard agents in these environments, suggesting that the added robustness increases generality and can adapt to non-stationary environments. Codes: https://github.com/adipandas/gym_multirotor

Episodic control enables sample efficiency in reinforcement learning by recalling past experiences from an episodic memory. We propose a new model-based episodic memory of trajectories addressing current limitations of episodic control. Our memory estimates trajectory values, guiding the agent towards good policies. Built upon the memory, we construct a complementary learning model via a dynamic hybrid control unifying model-based, episodic and habitual learning into a single architecture. Experiments demonstrate that our model allows significantly faster and better learning than other strong reinforcement learning agents across a variety of environments including stochastic and non-Markovian settings.