Deep reinforcement learning (RL) algorithms can use high-capacity deep networks to learn directly from image observations. However, these kinds of observation spaces present a number of challenges in practice, since the policy must now solve two problems: a representation learning problem, and a task learning problem. In this paper, we aim to explicitly learn representations that can accelerate reinforcement learning from images. We propose the stochastic latent actor-critic (SLAC) algorithm: a sample-efficient and high-performing RL algorithm for learning policies for complex continuous control tasks directly from high-dimensional image inputs. SLAC learns a compact latent representation space using a stochastic sequential latent variable model, and then learns a critic model within this latent space. By learning a critic within a compact state space, SLAC can learn much more efficiently than standard RL methods. The proposed model improves performance substantially over alternative representations as well, such as variational autoencoders. In fact, our experimental evaluation demonstrates that the sample efficiency of our resulting method is comparable to that of model-based RL methods that directly use a similar type of model for control. Furthermore, our method outperforms both model-free and model-based alternatives in terms of final performance and sample efficiency, on a range of difficult image-based control tasks.

Transfer learning across different reinforcement learning (RL) tasks is becoming an increasingly valuable area of research. We consider a goal-based multi-task RL framework and mechanisms by which previously solved tasks can reduce sample complexity and regret when the agent is faced with a new task. Specifically, we introduce two metrics on the state space that encode notions of traversibility of the state space for an agent. Using these metrics a topological covering is constructed by way of a set of landmark states in a fully self-supervised manner. We show that these landmark coverings confer theoretical advantages for transfer learning within the goal-based multi-task RL setting. Specifically, we demonstrate three mechanisms by which landmark coverings can be used for successful transfer learning. First, we extend the Landmark Options Via Reflection (LOVR) framework to this new topological covering; second, we use the landmark-centric value functions themselves as features and define a greedy zombie policy that achieves near oracle performance on a sequence of zero-shot transfer tasks; finally, motivated by the second transfer mechanism, we introduce a learned reward function that provides a more dense reward signal for goal-based RL. Our novel topological landmark covering confers beneficial theoretical results, bounding the Q values at each state-action pair. In doing so, we introduce a mechanism that performs action-pruning at infeasible actions which cannot possibly be part of an optimal policy for the current goal.

When deploying autonomous agents in the real world, we need effective ways of communicating objectives to them. Traditional skill learning has revolved around reinforcement and imitation learning, each with rigid constraints on the format of information exchanged between the human and the agent. While scalar rewards carry little information, demonstrations require significant effort to provide and may carry more information than is necessary. Furthermore, rewards and demonstrations are often defined and collected before training begins, when the human is most uncertain about what information would help the agent. In contrast, when humans communicate objectives with each other, they make use of a large vocabulary of informative behaviors, including non-verbal communication, and often communicate throughout learning, responding to observed behavior. In this way, humans communicate intent with minimal effort. In this paper, we propose such interactive learning as an alternative to reward or demonstration-driven learning. To accomplish this, we introduce a multi-agent training framework that enables an agent to learn from another agent who knows the current task. Through a series of experiments, we demonstrate the emergence of a variety of interactive learning behaviors, including information-sharing, information-seeking, and question-answering. Most importantly, we find that our approach produces an agent that is capable of learning interactively from a human user, without a set of explicit demonstrations or a reward function, and achieving significantly better performance cooperatively with a human than a human performing the task alone.

In many real-world scenarios, an autonomous agent often encounters various tasks within a single complex environment. We propose to build a graph abstraction over the environment structure to accelerate the learning of these tasks. Here, nodes are important points of interest (pivotal states) and edges represent feasible traversals between them. Our approach has two stages. First, we jointly train a latent pivotal state model and a curiosity-driven goal-conditioned policy in a task-agnostic manner. Second, provided with the information from the world graph, a high-level Manager quickly finds solution to new tasks and expresses subgoals in reference to pivotal states to a low-level Worker. The Worker can then also leverage the graph to easily traverse to the pivotal states of interest, even across long distance, and explore non-locally. We perform a thorough ablation study to evaluate our approach on a suite of challenging maze tasks, demonstrating significant advantages from the proposed framework over baselines that lack world graph knowledge in terms of performance and efficiency.

Developing agents that can quickly adapt their behavior to new tasks remains a challenge. Meta-learning has been applied to this problem, but previous methods require either specifying a reward function which can be tedious or providing demonstrations which can be inefficient. In this paper, we investigate if, and how, a "helper" agent can be trained to interactively adapt their behavior to maximize the reward of another agent, whom we call the "prime" agent, without observing their reward or receiving explicit demonstrations. To this end, we propose to meta-learn a helper agent along with a prime agent, who, during training, observes the reward function and serves as a surrogate for a human prime. We introduce a distribution of multi-agent cooperative foraging tasks, in which only the prime agent knows the objects that should be collected. We demonstrate that, from the emerged physical communication, the trained helper rapidly infers and collects the correct objects.

Current reinforcement learning methods fail if the reward function is imperfect, i.e. if the agent observes reward different from what it actually receives. We study this problem within the formalism of Corrupt Reward Markov Decision Processes (CRMDPs). We show that if the reward corruption in a CRMDP is sufficiently "spiky", the environment is solvable. We fully characterize the regret bound of a Spiky CRMDP, and introduce an algorithm that is able to detect its corrupt states. We show that this algorithm can be used to learn the optimal policy with any common reinforcement learning algorithm. Finally, we investigate our algorithm in a pair of simple gridworld environments, finding that our algorithm can detect the corrupt states and learn the optimal policy despite the corruption.

Being a complex subject of major importance in AI Safety research, value alignment has been studied from various perspectives in the last years. However, no final consensus on the design of ethical utility functions facilitating AI value alignment has been achieved yet. Given the urgency to identify systematic solutions, we postulate that it might be useful to start with the simple fact that for the utility function of an AI not to violate human ethical intuitions, it trivially has to be a model of these intuitions and reflect their variety $ - $ whereby the most accurate models pertaining to human entities being biological organisms equipped with a brain constructing concepts like moral judgements, are scientific models. Thus, in order to better assess the variety of human morality, we perform a transdisciplinary analysis applying a security mindset to the issue and summarizing variety-relevant background knowledge from neuroscience and psychology. We complement this information by linking it to augmented utilitarianism as a suitable ethical framework. Based on that, we propose first practical guidelines for the design of approximate ethical goal functions that might better capture the variety of human moral judgements. Finally, we conclude and address future possible challenges.

Recently, machine learning has prevailed in many academia and industrial applications. At the same time, quantum computing, once seen as not realizable, has been brought to markets by several tech giants. However, these machines are not fault-tolerant and can not execute very deep circuits. Therefore, it is urgent to design suitable algorithms and applications implementable on these machines. In this work, we demonstrate a novel approach which applies variational quantum circuits to deep reinforcement learning. With the proposed method, we can implement famous deep reinforcement learning algorithms such as experience replay and target network with variational quantum circuits. In this framework, with appropriate information encoding scheme, the possible quantum advantage is the number of circuit parameters with $poly(\log{} N)$ compared to $poly(N)$ in conventional neural network where $N$ is the dimension of input vectors. Such an approach can be deployed on near-term noisy intermediate-scale quantum machines.

Ludii is a general game system being developed as part of the ERC-funded Digital Ludeme Project (DLP). While its primary aim is to model, play, and analyse the full range of traditional strategy games, Ludii also has the potential to support a wide range of AI research topics and competitions. This paper describes some of the future competitions and challenges that we intend to run using the Ludii system, highlighting some of its most important aspects that can potentially lead to many algorithm improvements and new avenues of research. We compare and contrast our proposed competition motivations, goals and frameworks against those of existing general game playing competitions, addressing the strengths and weaknesses of each platform.

In this paper, the trajectory optimization problem for a multi-aerial base station (ABS) communication network is investigated. The objective is to find the trajectory of the ABSs so that the sum-rate of the users served by each ABS is maximized. To reach this goal, along with the optimal trajectory design, optimal power and sub-channel allocation is also of great importance to support the users with the highest possible data rates. To solve this complicated problem, we divide it into two sub-problems: ABS trajectory optimization sub-problem, and joint power and sub-channel assignment sub-problem. Then, based on the Q-learning method, we develop a distributed algorithm which solves these sub-problems efficiently, and does not need significant amount of information exchange between the ABSs and the core network. Simulation results show that although Q-learning is a model-free reinforcement learning technique, it has a remarkable capability to train the ABSs to optimize their trajectories based on the received reward signals, which carry decent information from the topology of the network.