We introduce inverse reinforcement learning (IRL) as an effective paradigm for training abstractive summarization models, imitating human summarization behaviors. Our IRL model estimates the reward function using a suite of important sub-rewards for summarization and concurrently optimizes the policy network. Experimental results across datasets in different domains (CNN/DailyMail and WikiHow) and various model sizes (BART-base and BART-large) demonstrate the superiority of our proposed IRL model for summarization over MLE and RL baselines. The resulting summaries exhibit greater similarity to human-crafted gold references, outperforming MLE and RL baselines on metrics such as ROUGE, coverage, novelty, compression ratio, factuality, and human evaluations.

Adopting reasonable strategies is challenging but crucial for an intelligent agent with limited resources working in hazardous, unstructured, and dynamic environments to improve the system's utility, decrease the overall cost, and increase mission success probability. This paper proposes a novel directed acyclic strategy graph decomposition approach based on Bayesian chaining to separate an intricate policy into several simple sub-policies and organize their relationships as Bayesian strategy networks (BSN). We integrate this approach into the state-of-the-art DRL method -- soft actor-critic (SAC), and build the corresponding Bayesian soft actor-critic (BSAC) model by organizing several sub-policies as a joint policy. We compare our method against the state-of-the-art deep reinforcement learning algorithms on the standard continuous control benchmarks in the OpenAI Gym environment. The results demonstrate that the promising potential of the BSAC method significantly improves training efficiency.

Nowadays, data are richly accessible to accumulate, and the increasingly powerful computing capability offers reasonable ease of handling big data. This remarkable scenario leads to a new way of solving some control problems that were previously challenging to analyze and solve. This paper proposes a new control approach, namely control with patterns (CWP), to handle data sets corresponding to nonlinear dynamical systems, where the feature abstraction must be considered for unstructured data feedback. For data sets of this kind, a new definition, namely exponential attraction on data sets, is proposed to describe nonlinear dynamical systems under consideration. Based on the data sets and parameterized Lyapunov functions, the problem for exponential attraction on data sets is converted to a pattern classification one. Furthermore, D-learning is proposed to perform CWP without knowledge of the system dynamics.

Contemporary artificial intelligence systems exhibit rapidly growing abilities accompanied by the growth of required resources, expansive datasets and corresponding investments into computing infrastructure. Although earlier successes predominantly focus on constrained settings, recent strides in fundamental research and applications aspire to create increasingly general systems. This evolving landscape presents a dual panorama of opportunities and challenges in refining the generalisation and transfer of knowledge - the extraction from existing sources and adaptation as a comprehensive foundation for tackling new problems. Within the domain of reinforcement learning (RL), the representation of knowledge manifests through various modalities, including dynamics and reward models, value functions, policies, and the original data. This taxonomy systematically targets these modalities and frames its discussion based on their inherent properties and alignment with different objectives and mechanisms for transfer. Where possible, we aim to provide coarse guidance delineating approaches which address requirements such as limiting environment interactions, maximising computational efficiency, and enhancing generalisation across varying axes of change. Finally, we analyse reasons contributing to the prevalence or scarcity of specific forms of transfer, the inherent potential behind pushing these frontiers, and underscore the significance of transitioning from designed to learned transfer.

We propose regularity as a novel reward signal for intrinsically-motivated reinforcement learning. Taking inspiration from child development, we postulate that striving for structure and order helps guide exploration towards a subspace of tasks that are not favored by naive uncertainty-based intrinsic rewards. Our generalized formulation of Regularity as Intrinsic Reward (RaIR) allows us to operationalize it within model-based reinforcement learning. In a synthetic environment, we showcase the plethora of structured patterns that can emerge from pursuing this regularity objective. We also demonstrate the strength of our method in a multiobject robotic manipulation environment. We incorporate RaIR into free play and use it to complement the model's epistemic uncertainty as an intrinsic reward. Doing so, we witness the autonomous construction of towers and other regular structures during free play, which leads to a substantial improvement in zero-shot downstream task performance on assembly tasks. Code and videos are available at https://sites.google.com/view/rair-project. Figure 1: Regularity as intrinsic reward yields ordered and symmetric patterns.

Optimisation problems, particularly combinatorial optimisation problems, are difficult to solve due to their complexity and hardness. Such problems have been successfully solved by evolutionary and swarm intelligence algorithms, especially in binary format. However, the approximation may suffer due to the the issues in balance between exploration and exploitation activities (EvE), which remain as the major challenge in this context. Although the complementary usage of multiple operators is becoming more popular for managing EvE with adaptive operator selection schemes, a bespoke adaptive selection system is still an important topic in research. Reinforcement Learning (RL) has recently been proposed as a way to customise and shape up a highly effective adaptive selection system. However, it is still challenging to handle the problem in terms of scalability. This paper proposes and assesses a RL-based novel approach to help develop a generalised framework for gaining, processing, and utilising the experiences for both the immediate and future use. The experimental results support the proposed approach with a certain level of success.

Replicating human-like dexterity in robot hands represents one of the largest open problems in robotics. Reinforcement learning is a promising approach that has achieved impressive progress in the last few years; however, the class of problems it has typically addressed corresponds to a rather narrow definition of dexterity as compared to human capabilities. To address this gap, we investigate piano-playing, a skill that challenges even the human limits of dexterity, as a means to test high-dimensional control, and which requires high spatial and temporal precision, and complex finger coordination and planning. We introduce RoboPianist, a system that enables simulated anthropomorphic hands to learn an extensive repertoire of 150 piano pieces where traditional model-based optimization struggles. We additionally introduce an open-sourced environment, benchmark of tasks, interpretable evaluation metrics, and open challenges for future study. Our website featuring videos, code, and datasets is available at https://kzakka.com/robopianist/

Interactive Intelligent Tutoring Systems (ITSs) enhance traditional ITSs by promoting effective learning through interactions and problem resolution in online education. Yet, proactive engagement, prioritizing resource optimization with planning and assessment capabilities, is often overlooked in current ITS designs. In this work, we investigate a new task, named Goal-oriented Intelligent Tutoring Systems (GITS), which aims to enable the student's mastery of a designated concept by strategically planning a customized sequence of exercises and assessment. To address the problem of goal-oriented policy learning in GITS, we propose a novel graph-based reinforcement learning framework, named Planning-Assessment-Interaction (PAI). Specifically, we first leverage cognitive structure information to improve state representation learning and action selection for planning the next action, which can be either to tutor an exercise or to assess the target concept. Further, we use a dynamically updated cognitive diagnosis model to simulate student responses to exercises and concepts. Three benchmark datasets across different subjects are constructed for enabling offline academic research on GITS. Experimental results demonstrate the effectiveness and efficiency of PAI and extensive analyses of various types of students are conducted to showcase the challenges in this task.

In this paper we investigate the use of reinforcement-learning based prediction approaches for a real drinking-water treatment plant. Developing such a prediction system is a critical step on the path to optimizing and automating water treatment. Before that, there are many questions to answer about the predictability of the data, suitable neural network architectures, how to overcome partial observability and more. We first describe this dataset, and highlight challenges with seasonality, nonstationarity, partial observability, and heterogeneity across sensors and operation modes of the plant. We then describe General Value Function (GVF) predictions -- discounted cumulative sums of observations -- and highlight why they might be preferable to classical n-step predictions common in time series prediction. We discuss how to use offline data to appropriately pre-train our temporal difference learning (TD) agents that learn these GVF predictions, including how to select hyperparameters for online fine-tuning in deployment. We find that the TD-prediction agent obtains an overall lower normalized mean-squared error than the n-step prediction agent. Finally, we show the importance of learning in deployment, by comparing a TD agent trained purely offline with no online updating to a TD agent that learns online. This final result is one of the first to motivate the importance of adapting predictions in real-time, for non-stationary high-volume systems in the real world.

The field of Multi-Agent Reinforcement Learning (MARL) is currently facing a reproducibility crisis. While solutions for standardized reporting have been proposed to address the issue, we still lack a benchmarking tool that enables standardization and reproducibility, while leveraging cutting-edge Reinforcement Learning (RL) implementations. In this paper, we introduce BenchMARL, the first MARL training library created to enable standardized benchmarking across different algorithms, models, and environments. BenchMARL uses TorchRL as its backend, granting it high performance and maintained state-of-the-art implementations while addressing the broad community of MARL PyTorch users. Its design enables systematic configuration and reporting, thus allowing users to create and run complex benchmarks from simple one-line inputs.