How to make imitation learning more general when demonstrations are relatively limited has been a persistent problem in reinforcement learning (RL). Poor demonstrations lead to narrow and biased date distribution, non-Markovian human expert demonstration makes it difficult for the agent to learn, and over-reliance on sub-optimal trajectories can make it hard for the agent to improve its performance. To solve these problems we propose a new algorithm named TD3fG that can smoothly transition from learning from experts to learning from experience. Our algorithm achieves good performance in the MUJOCO environment with limited and sub-optimal demonstrations. We use behavior cloning to train the network as a reference action generator and utilize it in terms of both loss function and exploration noise. This innovation can help agents extract a priori knowledge from demonstrations while reducing the detrimental effects of the poor Markovian properties of the demonstrations. It has a better performance compared to the BC+ fine-tuning and DDPGfD approach, especially when the demonstrations are relatively limited. We call our method TD3fG meaning TD3 from a generator.

Optimal control problems can be solved by first applying the Pontryagin maximum principle, followed by computing a solution of the corresponding unconstrained Hamiltonian dynamical system. In this paper, and to achieve a balance between robustness and efficiency, we learn a reduced Hamiltonian of the unconstrained Hamiltonian. This reduced Hamiltonian is learned by going backward in time and by minimizing the loss function resulting from application of the Pontryagin maximum principle's conditions. The robustness of our learning process is then further improved by progressively learning a posterior distribution of reduced Hamiltonians. This leads to a more efficient sampling of the generalized coordinates (position, velocity) of our phase space. Our solution framework applies to not only optimal control problems with finite-dimensional phase (state) spaces but also the infinite-dimensional case.

In a multi-task reinforcement learning setting, the learner commonly benefits from training on multiple related tasks by exploiting similarities among them. At the same time, the trained agent is able to solve a wider range of different problems. While this effect is well documented for model-free multi-task methods, we demonstrate a detrimental effect when using a single learned dynamics model for multiple tasks. Thus, we address the fundamental question of whether model-based multi-task reinforcement learning benefits from shared dynamics models in a similar way model-free methods do from shared policy networks. Using a single dynamics model, we see clear evidence of task confusion and reduced performance. As a remedy, enforcing an internal structure for the learned dynamics model by training isolated sub-networks for each task notably improves performance while using the same amount of parameters. We illustrate our findings by comparing both methods on a simple gridworld and a more complex vizdoom multi-task experiment.

Deep reinforcement learning (RL) presents a promising framework for learning impressive robot behaviors [1-4]. Yet, learning a complex long-horizon task using a single control policy is still challenging mainly due to its high computational costs and the exploration burdens of RL models [5]. A more practical solution is to decompose a whole task into smaller chunks of subtasks, learn a policy for each subtask, and sequentially execute the subtasks to accomplish the entire task [6-9]. However, naively executing one policy after another would fail when the subtask policy encounters a starting state never seen during training [6, 7, 9]. In other words, a terminal state of one subtask may fall outside of the set of starting states that the next subtask policy can handle, and thus fail to accomplish the subtask, as illustrated in Figure 1a. Especially in robot manipulation, complex interactions between a high-DoF robot and multiple objects could lead to a wide range of robot and object configurations, which are infeasible to be covered by a single policy [10]. Therefore, skill chaining with policies with limited capability is not trivial and requires adapting the policies to make them suitable for sequential execution. To resolve the mismatch between the terminal state distribution (i.e.

Applications of Reinforcement Learning (RL) in robotics are often limited by high data demand. On the other hand, approximate models are readily available in many robotics scenarios, making model-based approaches like planning a data-efficient alternative. Still, the performance of these methods suffers if the model is imprecise or wrong. In this sense, the respective strengths and weaknesses of RL and model-based planners are. In the present work, we investigate how both approaches can be integrated into one framework that combines their strengths. We introduce Learning to Execute (L2E), which leverages information contained in approximate plans to learn universal policies that are conditioned on plans. In our robotic manipulation experiments, L2E exhibits increased performance when compared to pure RL, pure planning, or baseline methods combining learning and planning.

Learning representations for pixel-based control has garnered significant attention recently in reinforcement learning. A wide range of methods have been proposed to enable efficient learning, leading to sample complexities similar to those in the full state setting. However, moving beyond carefully curated pixel data sets (centered crop, appropriate lighting, clear background, etc.) remains challenging. In this paper, we adopt a more difficult setting, incorporating background distractors, as a first step towards addressing this challenge. We present a simple baseline approach that can learn meaningful representations with no metric-based learning, no data augmentations, no world-model learning, and no contrastive learning. We then analyze when and why previously proposed methods are likely to fail or reduce to the same performance as the baseline in this harder setting and why we should think carefully about extending such methods beyond the well curated environments. Our results show that finer categorization of benchmarks on the basis of characteristics like density of reward, planning horizon of the problem, presence of task-irrelevant components, etc., is crucial in evaluating algorithms. Based on these observations, we propose different metrics to consider when evaluating an algorithm on benchmark tasks. We hope such a data-centric view can motivate researchers to rethink representation learning when investigating how to best apply RL to real-world tasks.

With breakthrough of AlphaGo, AI in human-computer game has become a very hot topic attracting researchers all around the world, which usually serves as an effective standard for testing artificial intelligence. Various game AI systems (AIs) have been developed such as Libratus, OpenAI Five and AlphaStar, beating professional human players. In this paper, we survey recent successful game AIs, covering board game AIs, card game AIs, first-person shooting game AIs and real time strategy game AIs. Through this survey, we 1) compare the main difficulties among different kinds of games for the intelligent decision making field ; 2) illustrate the mainstream frameworks and techniques for developing professional level AIs; 3) raise the challenges or drawbacks in the current AIs for intelligent decision making; and 4) try to propose future trends in the games and intelligent decision making techniques. Finally, we hope this brief review can provide an introduction for beginners, inspire insights for researchers in the filed of AI in games.

When I was in graduate school in the 1990s, one of my favorite classes was neural networks. Back then, we didn't have access to TensorFlow, PyTorch, or Keras; we programmed neurons, neural networks, and learning algorithms by hand with the formulas from textbooks. We didn't have access to cloud computing, and we coded sequential experiments that often ran overnight. There weren't platforms like Alteryx, Dataiku, SageMaker, or SAS to enable a machine learning proof of concept or manage the end-to-end MLops lifecycles. I was most interested in reinforcement learning algorithms, and I recall writing hundreds of reward functions to stabilise an inverted pendulum.

The feasibility of making profitable trades on a single asset on stock exchanges based on patterns identification has long attracted researchers. Reinforcement Learning (RL) and Natural Language Processing have gained notoriety in these single-asset trading tasks, but only a few works have explored their combination. Moreover, some issues are still not addressed, such as extracting market sentiment momentum through the explicit capture of sentiment features that reflect the market condition over time and assessing the consistency and stability of RL results in different situations. Filling this gap, we propose the Sentiment-Aware RL (SentARL) intelligent trading system that improves profit stability by leveraging market mood through an adaptive amount of past sentiment features drawn from textual news. We evaluated SentARL across twenty assets, two transaction costs, and five different periods and initializations to show its consistent effectiveness against baselines. Subsequently, this thorough assessment allowed us to identify the boundary between news coverage and market sentiment regarding the correlation of price-time series above which SentARL's effectiveness is outstanding.

The debate on whether or not humans have free will has been raging for centuries. Although there are good arguments based on our current understanding of the laws of nature for the view that it is not possible for humans to have free will, most people believe they do. This discrepancy begs for an explanation. If we accept that we do not have free will, we are faced with two problems: (1) while freedom is a very commonly used concept that everyone intuitively understands, what are we actually referring to when we say that an action or choice is "free" or not? And, (2) why is the belief in free will so common? Where does this belief come from, and what is its purpose, if any? In this paper, we examine these questions from the perspective of reinforcement learning (RL). RL is a framework originally developed for training artificial intelligence agents. However, it can also be used as a computational model of human decision making and learning, and by doing so, we propose that the first problem can be answered by observing that people's common sense understanding of freedom is closely related to the information entropy of an RL agent's normalized action values, while the second can be explained by the necessity for agents to model themselves as if they could have taken decisions other than those they actually took, when dealing with the temporal credit assignment problem. Put simply, we suggest that by applying the RL framework as a model for human learning it becomes evident that in order for us to learn efficiently and be intelligent we need to view ourselves as if we have free will.