Learning optimal policies in sparse rewards settings is difficult as the learning agent has little to no feedback on the quality of its actions. In these situations, a good strategy is to focus on exploration, hopefully leading to the discovery of a reward signal to improve on. A learning algorithm capable of dealing with this kind of settings has to be able to (1) explore possible agent behaviors and (2) exploit any possible discovered reward. Efficient exploration algorithms have been proposed that require to define a behavior space, that associates to an agent its resulting behavior in a space that is known to be worth exploring. The need to define this space is a limitation of these algorithms. In this work, we introduce STAX, an algorithm designed to learn a behavior space on-the-fly and to explore it while efficiently optimizing any reward discovered. It does so by separating the exploration and learning of the behavior space from the exploitation of the reward through an alternating two-steps process. In the first step, STAX builds a repertoire of diverse policies while learning a low-dimensional representation of the high-dimensional observations generated during the policies evaluation. In the exploitation step, emitters are used to optimize the performance of the discovered rewarding solutions. Experiments conducted on three different sparse reward environments show that STAX performs comparably to existing baselines while requiring much less prior information about the task as it autonomously builds the behavior space.

We propose a novel approach to explainable AI (XAI) based on the concept of "instruction" from neural networks. In this case study, we demonstrate how a superhuman neural network might instruct human trainees as an alternative to traditional approaches to XAI. Specifically, an AI examines human actions and calculates variations on the human strategy that lead to better performance. Experiments with a JHU/APL-developed AI player for the cooperative card game Hanabi suggest this technique makes unique contributions to explainability while improving human performance. One area of focus for Instructive AI is in the significant discrepancies that can arise between a human's actual strategy and the strategy they profess to use. This inaccurate self-assessment presents a barrier for XAI, since explanations of an AI's strategy may not be properly understood or implemented by human recipients. We have developed and are testing a novel, Instructive AI approach that estimates human strategy by observing human actions. With neural networks, this allows a direct calculation of the changes in weights needed to improve the human strategy to better emulate a more successful AI. Subjected to constraints (e.g. sparsity) these weight changes can be interpreted as recommended changes to human strategy (e.g. "value A more, and value B less"). Instruction from AI such as this functions both to help humans perform better at tasks, but also to better understand, anticipate, and correct the actions of an AI. Results will be presented on AI instruction's ability to improve human decision-making and human-AI teaming in Hanabi.

Many possible fields of application of robots in real world settings hinge on the ability of robots to grasp objects. As a result, robot grasping has been an active field of research for many years. With our publication we contribute to the endeavor of enabling robots to grasp, with a particular focus on bin picking applications. Bin picking is especially challenging due to the often cluttered and unstructured arrangement of objects and the often limited graspability of objects by simple top down grasps. To tackle these challenges, we propose a fully self-supervised reinforcement learning approach based on a hybrid discrete-continuous adaptation of soft actor-critic (SAC). We employ parametrized motion primitives for pushing and grasping movements in order to enable a flexibly adaptable behavior to the difficult setups we consider. Furthermore, we use data augmentation to increase sample efficiency. We demonnstrate our proposed method on challenging picking scenarios in which planar grasp learning or action discretization methods would face a lot of difficulties

In spoken dialogue systems, we aim to deploy artificial intelligence to build automated dialogue agents that can converse with humans. Dialogue systems are increasingly being designed to move beyond just imitating conversation and also improve from such interactions over time. In this survey, we present a broad overview of methods developed to build dialogue systems over the years. Different use cases for dialogue systems ranging from task-based systems to open domain chatbots motivate and necessitate specific systems. Starting from simple rule-based systems, research has progressed towards increasingly complex architectures trained on a massive corpus of datasets, like deep learning systems. Motivated with the intuition of resembling human dialogues, progress has been made towards incorporating emotions into the natural language generator, using reinforcement learning. While we see a trend of highly marginal improvement on some metrics, we find that limited justification exists for the metrics, and evaluation practices are not uniform. To conclude, we flag these concerns and highlight possible research directions.

Reinforcement Learning (RL) has been witnessed its potential for training a dialogue policy agent towards maximizing the accumulated rewards given from users. However, the reward can be very sparse for it is usually only provided at the end of a dialog session, which causes unaffordable interaction requirements for an acceptable dialog agent. Distinguished from many efforts dedicated to optimizing the policy and recovering the reward alternatively which suffers from easily getting stuck in local optima and model collapse, we decompose the adversarial training into two steps: 1) we integrate a pre-trained language model as a discriminator to judge whether the current system action is good enough for the last user action (i.e., \textit{next action prediction}); 2) the discriminator gives and extra local dense reward to guide the agent's exploration. The experimental result demonstrates that our method significantly improves the complete rate (~4.4\%) and success rate (~8.0\%) of the dialogue system.

Autonomous 3D environment exploration is a fundamental task for various applications such as navigation. The goal of exploration is to investigate a new environment and build its occupancy map efficiently. In this paper, we propose a new method which grants an agent two intertwined options of behaviors: "look-around" and "frontier navigation". This is implemented by an option-critic architecture and trained by reinforcement learning algorithms. In each timestep, an agent produces an option and a corresponding action according to the policy. We also take advantage of macro-actions by incorporating classic path-planning techniques to increase training efficiency. We demonstrate the effectiveness of the proposed method on two publicly available 3D environment datasets and the results show our method achieves higher coverage than competing techniques with better efficiency.

We propose a deep reinforcement learning (RL) method to learn large neighborhood search (LNS) policy for integer programming (IP). The RL policy is trained as the destroy operator to select a subset of variables at each step, which is reoptimized by an IP solver as the repair operator. However, the combinatorial number of variable subsets prevents direct application of typical RL algorithms. To tackle this challenge, we represent all subsets by factorizing them into binary decisions on each variable. We then design a neural network to learn policies for each variable in parallel, trained by a customized actor-critic algorithm. We evaluate the proposed method on four representative IP problems. Results show that it can find better solutions than SCIP in much less time, and significantly outperform other LNS baselines with the same runtime. Moreover, these advantages notably persist when the policies generalize to larger problems. Further experiments with Gurobi also reveal that our method can outperform this state-of-the-art commercial solver within the same time limit.

Attacks exploiting human attentional vulnerability have posed severe threats to cybersecurity. In this work, we identify and formally define a new type of proactive attentional attacks called Informational Denial-of-Service (IDoS) attacks that generate a large volume of feint attacks to overload human operators and hide real attacks among feints. We incorporate human factors (e.g., levels of expertise, stress, and efficiency) and empirical results (e.g., the Yerkes-Dodson law and the sunk cost fallacy) to model the operators' attention dynamics and their decision-making processes along with the real-time alert monitoring and inspection. To assist human operators in timely and accurately dismissing the feints and escalating the real attacks, we develop a Resilient and Adaptive Data-driven alert and Attention Management Strategy (RADAMS) that de-emphasizes alerts selectively based on the alerts' observable features. RADAMS uses reinforcement learning to achieve a customized and transferable design for various human operators and evolving IDoS attacks. The integrated modeling and theoretical analysis lead to the Product Principle of Attention (PPoA), fundamental limits, and the tradeoff among crucial human and economic factors. Experimental results corroborate that the proposed strategy outperforms the default strategy and can reduce the IDoS risk by as much as 20%. Besides, the strategy is resilient to large variations of costs, attack frequencies, and human attention capacities. We have recognized interesting phenomena such as attentional risk equivalency, attacker's dilemma, and the half-truth optimal attack strategy.

Independent reinforcement learning algorithms have no theoretical guarantees for finding the best policy in multi-agent settings. However, in practice, prior works have reported good performance with independent algorithms in some domains and bad performance in others. Moreover, a comprehensive study of the strengths and weaknesses of independent algorithms is lacking in the literature. In this paper, we carry out an empirical comparison of the performance of independent algorithms on four PettingZoo environments that span the three main categories of multi-agent environments, i.e., cooperative, competitive, and mixed. We show that in fully-observable environments, independent algorithms can perform on par with multi-agent algorithms in cooperative and competitive settings. For the mixed environments, we show that agents trained via independent algorithms learn to perform well individually, but fail to learn to cooperate with allies and compete with enemies. We also show that adding recurrence improves the learning of independent algorithms in cooperative partially observable environments.

Reward is the driving force for reinforcement-learning agents. This paper is dedicated to understanding the expressivity of reward as a way to capture tasks that we would want an agent to perform. We frame this study around three new abstract notions of "task" that might be desirable: (1) a set of acceptable behaviors, (2) a partial ordering over behaviors, or (3) a partial ordering over trajectories. Our main results prove that while reward can express many of these tasks, there exist instances of each task type that no Markov reward function can capture. We then provide a set of polynomial-time algorithms that construct a Markov reward function that allows an agent to optimize tasks of each of these three types, and correctly determine when no such reward function exists. We conclude with an empirical study that corroborates and illustrates our theoretical findings.