Duo Xu

Duo Xu, Faramarz Fekri

In this study, we address the challenge of learning generalizable policies for compositional tasks defined by logical specifications. These tasks consist of multiple temporally extended sub-tasks. Due to the sub-task inter-dependencies and sparse reward issue in long-horizon tasks, existing reinforcement learning (RL) approaches, such as task-conditioned and goal-conditioned policies, continue to struggle with slow convergence and sub-optimal performance in generalizing to compositional tasks. To overcome these limitations, we introduce a new hierarchical RL framework that enhances the efficiency and optimality of task generalization. At the high level, we present an implicit planner specifically designed for generalizing compositional tasks. This planner selects the next sub-task and estimates the multi-step return for completing the remaining task to complete from the current state. It learns a latent transition model and performs planning in the latent space by using a graph neural network (GNN). Subsequently, the high-level planner's selected sub-task guides the low-level agent to effectively handle long-horizon tasks, while the multi-step return encourages the low-level policy to account for future sub-task dependencies, enhancing its optimality. We conduct comprehensive experiments to demonstrate the framework's advantages over previous methods in terms of both efficiency and optimality.

Duo Xu, Faramarz Fekri

Recently deep reinforcement learning has achieved tremendous success in wide ranges of applications. However, it notoriously lacks data-efficiency and interpretability. Data-efficiency is important as interacting with the environment is expensive. Further, interpretability can increase the transparency of the black-box-style deep RL models and hence gain trust from the users. In this work, we propose a new hierarchical framework via symbolic RL, leveraging a symbolic transition model to improve the data-efficiency and introduce the interpretability for learned policy. This framework consists of a high-level agent, a subtask solver and a symbolic transition model. Without assuming any prior knowledge on the state transition, we adopt inductive logic programming (ILP) to learn the rules of symbolic state transitions, introducing interpretability and making the learned behavior understandable to users. In empirical experiments, we confirmed that the proposed framework offers approximately between 30\% to 40\% more data efficiency over previous methods.

Duo Xu, Mohit Agarwal, Ekansh Gupta et al.

Providing Reinforcement Learning (RL) agents with human feedback can dramatically improve various aspects of learning. However, previous methods require human observer to give inputs explicitly (e.g., press buttons, voice interface), burdening the human in the loop of RL agent's learning process. Further, it is sometimes difficult or impossible to obtain the explicit human advise (feedback), e.g., autonomous driving, disabled rehabilitation, etc. In this work, we investigate capturing human's intrinsic reactions as implicit (and natural) feedback through EEG in the form of error-related potentials (ErrP), providing a natural and direct way for humans to improve the RL agent learning. As such, the human intelligence can be integrated via implicit feedback with RL algorithms to accelerate the learning of RL agent. We develop three reasonably complex 2D discrete navigational games to experimentally evaluate the overall performance of the proposed work. Major contributions of our work are as follows, (i) we propose and experimentally validate the zero-shot learning of ErrPs, where the ErrPs can be learned for one game, and transferred to other unseen games, (ii) we propose a novel RL framework for integrating implicit human feedbacks via ErrPs with RL agent, improving the label efficiency and robustness to human mistakes, and (iii) compared to prior works, we scale the application of ErrPs to reasonably complex environments, and demonstrate the significance of our approach for accelerated learning through real user experiments.

Duo Xu

State space models (SSM) have been widely applied for the analysis and visualization of large sequential datasets. Sequential Monte Carlo (SMC) is a very popular particle-based method to sample latent states from intractable posteriors. However, SSM is significantly influenced by the choice of the proposal. Recently Hamiltonian Monte Carlo (HMC) sampling has shown success in many practical problems. In this paper, we propose an SMC augmented by HMC (HSMC) for inference and model learning of nonlinear SSM, which can exempt us from learning proposals and reduce the model complexity significantly. Based on the measure preserving property of HMC, the particles directly generated by transition function can approximate the posterior of latent states arbitrarily well. In order to better adapt to the local geometry of latent space, the HMC is conducted on Riemannian manifold defined by a positive definite metric. In addition, we show that the proposed HSMC method can improve SSMs realized by both Gaussian Processes (GP) and Neural Network (NN).