Fully cooperative, partially observable multi-agent problems are ubiquitous in the real world. In this paper, we focus on a specific subclass of coordination problems in which humans are able to discover self-explaining deviations (SEDs). SEDs are actions that deviate from the common understanding of what reasonable behavior would be in normal circumstances. They are taken with the intention of causing another agent or other agents to realize, using theory of mind, that the circumstance must be abnormal. We first motivate SED with a real world example and formalize its definition. Next, we introduce a novel algorithm, improvement maximizing self-explaining deviations (IMPROVISED), to perform SEDs. Lastly, we evaluate IMPROVISED both in an illustrative toy setting and the popular benchmark setting Hanabi, where it is the first method to produce so called finesse plays, which are regarded as one of the more iconic examples of human theory of mind.

We develop a Reinforcement Learning (RL) framework for improving an existing behavior policy via sparse, user-interpretable changes. Our goal is to make minimal changes while gaining as much benefit as possible. We define a minimal change as having a sparse, global contrastive explanation between the original and proposed policy. We improve the current policy with the constraint of keeping that global contrastive explanation short. We demonstrate our framework with a discrete MDP and a continuous 2D navigation domain.

Robots are increasingly expected to manipulate objects in ever more unstructured environments where the object properties have high perceptual uncertainty from any single sensory modality. This directly impacts successful object manipulation. In this work, we propose a reinforcement learning-based motion planning framework for object manipulation which makes use of both on-the-fly multisensory feedback and a learned attention-guided deep affordance model as perceptual states. The affordance model is learned from multiple sensory modalities, including vision and touch (tactile and force/torque), which is designed to predict and indicate the manipulable regions of multiple affordances (i.e., graspability and pushability) for objects with similar appearances but different intrinsic properties (e.g., mass distribution). A DQN-based deep reinforcement learning algorithm is then trained to select the optimal action for successful object manipulation. To validate the performance of the proposed framework, our method is evaluated and benchmarked using both an open dataset and our collected dataset. The results show that the proposed method and overall framework outperform existing methods and achieve better accuracy and higher efficiency.

Motion control algorithms in the presence of pedestrians are critical for the development of safe and reliable Autonomous Vehicles (AVs). Traditional motion control algorithms rely on manually designed decision-making policies which neglect the mutual interactions between AVs and pedestrians. On the other hand, recent advances in Deep Reinforcement Learning allow for the automatic learning of policies without manual designs. To tackle the problem of decision-making in the presence of pedestrians, the authors introduce a framework based on Social Value Orientation and Deep Reinforcement Learning (DRL) that is capable of generating decision-making policies with different driving styles. The policy is trained using state-of-the-art DRL algorithms in a simulated environment. A novel computationally-efficient pedestrian model that is suitable for DRL training is also introduced. We perform experiments to validate our framework and we conduct a comparative analysis of the policies obtained with two different model-free Deep Reinforcement Learning Algorithms. Simulations results show how the developed model exhibits natural driving behaviours, such as short-stopping, to facilitate the pedestrian's crossing.

In this paper, we study the problem of global reward maximization with only partial distributed feedback. This problem is motivated by several real-world applications (e.g., cellular network configuration, dynamic pricing, and policy selection) where an action taken by a central entity influences a large population that contributes to the global reward. However, collecting such reward feedback from the entire population not only incurs a prohibitively high cost but often leads to privacy concerns. To tackle this problem, we consider differentially private distributed linear bandits, where only a subset of users from the population are selected (called clients) to participate in the learning process and the central server learns the global model from such partial feedback by iteratively aggregating these clients' local feedback in a differentially private fashion. We then propose a unified algorithmic learning framework, called differentially private distributed phased elimination (DP-DPE), which can be naturally integrated with popular differential privacy (DP) models (including central DP, local DP, and shuffle DP). Furthermore, we prove that DP-DPE achieves both sublinear regret and sublinear communication cost. Interestingly, DP-DPE also achieves privacy protection "for free" in the sense that the additional cost due to privacy guarantees is a lower-order additive term. In addition, as a by-product of our techniques, the same results of "free" privacy can also be achieved for the standard differentially private linear bandits. Finally, we conduct simulations to corroborate our theoretical results and demonstrate the effectiveness of DP-DPE.

Optimistic algorithms have been extensively studied for regret minimization in episodic tabular MDPs, both from a minimax and an instance-dependent view. However, for the PAC RL problem, where the goal is to identify a near-optimal policy with high probability, little is known about their instance-dependent sample complexity. A negative result of Wagenmaker et al. (2021) suggests that optimistic sampling rules cannot be used to attain the (still elusive) optimal instance-dependent sample complexity. On the positive side, we provide the first instance-dependent bound for an optimistic algorithm for PAC RL, BPI-UCRL, for which only minimax guarantees were available (Kaufmann et al., 2021). While our bound features some minimal visitation probabilities, it also features a refined notion of sub-optimality gap compared to the value gaps that appear in prior work. Moreover, in MDPs with deterministic transitions, we show that BPI-UCRL is actually near-optimal. On the technical side, our analysis is very simple thanks to a new "target trick" of independent interest. We complement these findings with a novel hardness result explaining why the instance-dependent complexity of PAC RL cannot be easily related to that of regret minimization, unlike in the minimax regime.

Teaching a deep reinforcement learning (RL) agent to follow instructions in multi-task environments is a challenging problem. We consider that user defines every task by a linear temporal logic (LTL) formula. However, some causal dependencies in complex environments may be unknown to the user in advance. Hence, when human user is specifying instructions, the robot cannot solve the tasks by simply following the given instructions. In this work, we propose a hierarchical reinforcement learning (HRL) framework in which a symbolic transition model is learned to efficiently produce high-level plans that can guide the agent efficiently solve different tasks. Specifically, the symbolic transition model is learned by inductive logic programming (ILP) to capture logic rules of state transitions. By planning over the product of the symbolic transition model and the automaton derived from the LTL formula, the agent can resolve causal dependencies and break a causally complex problem down into a sequence of simpler low-level sub-tasks. We evaluate the proposed framework on three environments in both discrete and continuous domains, showing advantages over previous representative methods.

We study representation learning for Offline Reinforcement Learning (RL), focusing on the important task of Offline Policy Evaluation (OPE). Recent work shows that, in contrast to supervised learning, realizability of the Q-function is not enough for learning it. Two sufficient conditions for sample-efficient OPE are Bellman completeness and coverage. Prior work often assumes that representations satisfying these conditions are given, with results being mostly theoretical in nature. In this work, we propose BCRL, which directly learns from data an approximately linear Bellman complete representation with good coverage. With this learned representation, we perform OPE using Least Square Policy Evaluation (LSPE) with linear functions in our learned representation. We present an end-to-end theoretical analysis, showing that our two-stage algorithm enjoys polynomial sample complexity provided some representation in the rich class considered is linear Bellman complete. Empirically, we extensively evaluate our algorithm on challenging, image-based continuous control tasks from the Deepmind Control Suite. We show our representation enables better OPE compared to previous representation learning methods developed for off-policy RL (e.g., CURL, SPR). BCRL achieve competitive OPE error with the state-of-the-art method Fitted Q-Evaluation (FQE), and beats FQE when evaluating beyond the initial state distribution. Our ablations show that both linear Bellman complete and coverage components of our method are crucial.

This primer is an attempt to provide a detailed summary of the different facets of lifelong learning. We start with Chapter 2 which provides a high-level overview of lifelong learning systems. In this chapter, we discuss prominent scenarios in lifelong learning (Section 2.4), provide 8 Introduction a high-level organization of different lifelong learning approaches (Section 2.5), enumerate the desiderata for an ideal lifelong learning system (Section 2.6), discuss how lifelong learning is related to other learning paradigms (Section 2.7), describe common metrics used to evaluate lifelong learning systems (Section 2.8). This chapter is more useful for readers who are new to lifelong learning and want to get introduced to the field without focusing on specific approaches or benchmarks. The remaining chapters focus on specific aspects (either learning algorithms or benchmarks) and are more useful for readers who are looking for specific approaches or benchmarks. Chapter 3 focuses on regularization-based approaches that do not assume access to any data from previous tasks. Chapter 4 discusses memory-based approaches that typically use a replay buffer or an episodic memory to save subset of data across different tasks. Chapter 5 focuses on different architecture families (and their instantiations) that have been proposed for training lifelong learning systems. Following these different classes of learning algorithms, we discuss the commonly used evaluation benchmarks and metrics for lifelong learning (Chapter 6) and wrap up with a discussion of future challenges and important research directions in Chapter 7.

Existing offline reinforcement learning (RL) methods face a few major challenges, particularly the distributional shift between the learned policy and the behavior policy. Offline Meta-RL is emerging as a promising approach to address these challenges, aiming to learn an informative meta-policy from a collection of tasks. Nevertheless, as shown in our empirical studies, offline Meta-RL could be outperformed by offline single-task RL methods on tasks with good quality of datasets, indicating that a right balance has to be delicately calibrated between "exploring" the out-of-distribution state-actions by following the meta-policy and "exploiting" the offline dataset by staying close to the behavior policy. Motivated by such empirical analysis, we explore model-based offline Meta-RL with regularized Policy Optimization (MerPO), which learns a meta-model for efficient task structure inference and an informative meta-policy for safe exploration of out-of-distribution state-actions. In particular, we devise a new meta-Regularized model-based Actor-Critic (RAC) method for within-task policy optimization, as a key building block of MerPO, using conservative policy evaluation and regularized policy improvement; and the intrinsic tradeoff therein is achieved via striking the right balance between two regularizers, one based on the behavior policy and the other on the meta-policy. We theoretically show that the learnt policy offers guaranteed improvement over both the behavior policy and the meta-policy, thus ensuring the performance improvement on new tasks via offline Meta-RL. Experiments corroborate the superior performance of MerPO over existing offline Meta-RL methods. Offline reinforcement learning (a.k.a., batch RL) has recently attracted extensive attention by learning from offline datasets previously collected via some behavior policy (Kumar et al., 2020). However, the performance of existing offline RL methods could degrade significantly due to the following issues: 1) the possibly poor quality of offline datasets (Levine et al., 2020) and 2) the inability to generalize to different environments (Li et al., 2020b). To tackle these challenges, offline Meta -RL (Li et al., 2020a; Dorfman & Tamar, 2020; Mitchell et al., 2020; Li et al., 2020b) has emerged very recently by leveraging the knowledge of similar offline RL tasks (Y u et al., 2021a).