Our deep reinforcement learning scheme provides solutions for known challenges in quantum technology and opens new avenues for the control of quantum devices. Expand abstract.

The design, accurate preparation and manipulation of quantum states are essential operational tasks at the heart of quantum technologies. Nowadays, physical parameters of quantum devices and networks can be controlled with unprecedented accuracy and flexibility. However, the generation of well-controlled current states is still a nagging bottleneck, especially when different circuit elements are integrated together. In this work, we show how machine learning can effectively address this challenge and outperform the current existing methods. To this end, we exploit deep reinforcement learning to prepare prescribed quantum current states within a short time scale and with a high fidelity. To highlight our method, we show how to engineer bosonic persistent currents in ring circuits as they are key ingredients in different quantum technology devices. With our approach, quantum current states characterized by a single winding number or entangled currents with two winding numbers can be prepared superseding the existing protocols. In addition, we generated quantum states entangling a larger set of different winding numbers. Our deep reinforcement learning scheme provides solutions for known challenges in quantum technology and opens new avenues for the control of quantum devices.

Overall, the virtual rodent facilitates grounded collaborations between deep reinforcement learning and motor neuroscience. Expand abstract.

Parallel developments in neuroscience and deep learning have led to mutually productive exchanges, pushing our understanding of real and artificial neural networks in sensory and cognitive systems. However, this interaction between fields is less developed in the study of motor control. In this work, we develop a virtual rodent as a platform for the grounded study of motor activity in artificial models of embodied control. We then use this platform to study motor activity across contexts by training a model to solve four complex tasks. Using methods familiar to neuroscientists, we describe the behavioral representations and algorithms employed by different layers of the network using a neuroethological approach to characterize motor activity relative to the rodent's behavior and goals. We find that the model uses two classes of representations which respectively encode the task-specific behavioral strategies and task-invariant behavioral kinematics. These representations are reflected in the sequential activity and population dynamics of neural subpopulations. Overall, the virtual rodent facilitates grounded collaborations between deep reinforcement learning and motor neuroscience.

Our end-to-end differentiable model is trained with A2C, a standard deep reinforcement learning algorithm. Expand abstract.

We focus on a class of real-world domains, where gathering hierarchical knowledge is required to accomplish a task. Many problems can be represented in this manner, such as network penetration testing, targeted advertising or medical diagnosis. In our formalization, the task is to sequentially request pieces of information about a sample to build the knowledge hierarchy and terminate when suitable. Any of the learned pieces of information can be further analyzed, resulting in a complex and variable action space. We present a combination of techniques in which the knowledge hierarchy is explicitly represented and given to a deep reinforcement learning algorithm as its input. To process the hierarchical input, we employ Hierarchical Multiple-Instance Learning and to cope with the complex action space, we factor it with hierarchical softmax. Our end-to-end differentiable model is trained with A2C, a standard deep reinforcement learning algorithm. We demonstrate the method in a set of seven classification domains, where the task is to achieve the best accuracy with a set budget on the amount of information retrieved. Compared to baseline algorithms, our method achieves not only better results, but also better generalization.

We show our learned policy successfully runs on a real robot without any fine tuning, thus bridging the reality gap. Expand abstract.

Using simulation to train robot manipulation policies holds the promise of an almost unlimited amount of training data, generated safely out of harm's way. One of the key challenges of using simulation, to date, has been to bridge the reality gap, so that policies trained in simulation can be deployed in the real world. We explore the reality gap in the context of learning a contextual policy for multi-fingered robotic grasping. We propose a Grasping Objects Approach for Tactile (GOAT) robotic hands, learning to overcome the reality gap problem. In our approach we use human hand motion demonstration to initialize and reduce the search space for learning. We contextualize our policy with the bounding cuboid dimensions of the object of interest, which allows the policy to work on a more flexible representation than directly using an image or point cloud. Leveraging fingertip touch sensors in the hand allows the policy to overcome the reduction in geometric information introduced by the coarse bounding box, as well as pose estimation uncertainty. We show our learned policy successfully runs on a real robot without any fine tuning, thus bridging the reality gap.

To tackle this challenging problem, this paper applies the Deep Reinforcement Learning (DRL) method by 1) representing the state by a coverage bitmap to capture the spatial correlation of GUs/ABSs, whose dimension and associated neural network complexity is invariant with arbitrarily large N; and 2) designing the action... Expand abstract.

Unmanned aerial vehicles (UAVs) can be utilized as aerial base stations (ABSs) to assist terrestrial infrastructure for keeping wireless connectivity in various emergency scenarios. To maximize the coverage rate of N ground users (GUs) by jointly placing multiple ABSs with limited coverage range is known to be a NP-hard problem with exponential complexity in N. The problem is further complicated when the coverage range becomes irregular due to site-specific blockage (e.g., buildings) on the air-ground channel in the 3-dimensional (3D) space. To tackle this challenging problem, this paper applies the Deep Reinforcement Learning (DRL) method by 1) representing the state by a coverage bitmap to capture the spatial correlation of GUs/ABSs, whose dimension and associated neural network complexity is invariant with arbitrarily large N; and 2) designing the action and reward for the DRL agent to effectively learn from the dynamic interactions with the complicated propagation environment represented by a 3D Terrain Map. Specifically, a novel two-level design approach is proposed, consisting of a preliminary design based on the dominant line-of-sight (LoS) channel model, and an advanced design to further refine the ABS positions based on site-specific LoS/non-LoS channel states. The double deep Q-network (DQN) with Prioritized Experience Replay (Prioritized Replay DDQN) algorithm is applied to train the policy of multi-ABS placement decision. Numerical results show that the proposed approach significantly improves the coverage rate in complex environment, compared to the benchmark DQN and K-means algorithms.

Learning good representations is a long standing problem in reinforcement learning (RL). Expand abstract.

Learning good representations is a long standing problem in reinforcement learning (RL). One of the conventional ways to achieve this goal in the supervised setting is through regularization of the parameters. Extending some of these ideas to the RL setting has not yielded similar improvements in learning. In this paper, we develop an online regularization framework for decorrelating features in RL and demonstrate its utility in several test environments. We prove that the proposed algorithm converges in the linear function approximation setting and does not change the main objective of maximizing cumulative reward. We demonstrate how to scale the approach to deep RL using the Gramian of the features achieving linear computational complexity in the number of features and squared complexity in size of the batch. We conduct an extensive empirical study of the new approach on Atari 2600 games and show a significant improvement in sample efficiency in 40 out of 49 games.

Balancing exploration and exploitation is a fundamental part of reinforcement learning, yet most state-of-the-art algorithms use a naive exploration protocol like $\epsilon$-greedy. Expand abstract.

Balancing exploration and exploitation is a fundamental part of reinforcement learning, yet most state-of-the-art algorithms use a naive exploration protocol like $\epsilon$-greedy. This contributes to the problem of high sample complexity, as the algorithm wastes effort by repeatedly visiting parts of the state space that have already been explored. We introduce a novel method based on Bayesian linear regression and latent space embedding to generate an intrinsic reward signal that encourages the learning agent to seek out unexplored parts of the state space. This method is computationally efficient, simple to implement, and can extend any state-of-the-art reinforcement learning algorithm. We evaluate the method on a range of algorithms and challenging control tasks, on both simulated and physical robots, demonstrating how the proposed method can significantly improve sample complexity.

In this paper, we extend maximum entropy inverse reinforcement learning to the N-player case in order to solve inverse dynamic games with continuous-valued state and control spaces. Expand abstract.

Dynamic game theory provides mathematical means for modeling the interaction between several players, where their decisions are explained by individual cost functions. The inverse problem of dynamic games, where cost functions are sought which explain observed behavior, has recently gained attention due to its potential application for identification of biological systems and the possibility of generalizing inverse optimal control results. In this paper, we extend maximum entropy inverse reinforcement learning to the N-player case in order to solve inverse dynamic games with continuous-valued state and control spaces. On this basis, we first present a method for identification of cost function parameters in a cooperative game. Afterwards, we propose an approach for identifying cost function parameters which explain the behavior of the players in a non-cooperative setting, i.e. open-loop and feedback Nash equilibrium behaviors. Furthermore, we give results on the unbiasedness of the estimation of cost function parameters for each class of inverse dynamic game. The applicability of the methods is demonstrated with simulation examples of a nonlinear and a linear-quadratic dynamic game.

Existing value-factorized based Multi-Agent deep Reinforce-ment Learning (MARL) approaches are well-performing invarious multi-agent cooperative environment under thecen-tralized training and decentralized execution(CTDE) scheme,where all agents are trained together by the centralized valuenetwork and each agent execute its policy... Expand abstract.

Existing value-factorized based Multi-Agent deep Reinforce-ment Learning (MARL) approaches are well-performing invarious multi-agent cooperative environment under thecen-tralized training and decentralized execution(CTDE) scheme,where all agents are trained together by the centralized valuenetwork and each agent execute its policy independently. How-ever, an issue remains open: in the centralized training process,when the environment for the team is partially observable ornon-stationary, i.e., the observation and action informationof all the agents cannot represent the global states, existingmethods perform poorly and sample inefficiently. Regret Min-imization (RM) can be a promising approach as it performswell in partially observable and fully competitive settings.However, it tends to model others as opponents and thus can-not work well under the CTDE scheme. In this work, wepropose a novel team RM based Bayesian MARL with threekey contributions: (a) we design a novel RM method to traincooperative agents as a team and obtain a team regret-basedpolicy for that team; (b) we introduce a novel method to de-compose the team regret to generate the policy for each agentfor decentralized execution; (c) to further improve the perfor-mance, we leverage a differential particle filter (a SequentialMonte Carlo method) network to get an accurate estimation ofthe state for each agent. Experimental results on two-step ma-trix games (cooperative game) and battle games (large-scalemixed cooperative-competitive games) demonstrate that ouralgorithm significantly outperforms state-of-the-art methods.

In order to prepare longitudinal, irregularly sampled, clinical datasets for reinforcement learning algorithms, many researchers will resample the time series data to short, regular intervals and use last-observation-carried-forward (LOCF) imputation to fill in these gaps. Expand abstract.

There is an emerging trend in the reinforcement learning for healthcare literature. In order to prepare longitudinal, irregularly sampled, clinical datasets for reinforcement learning algorithms, many researchers will resample the time series data to short, regular intervals and use last-observation-carried-forward (LOCF) imputation to fill in these gaps. Typically, they will not maintain any explicit information about which values were imputed. In this work, we (1) call attention to this practice and discuss its potential implications; (2) propose an alternative representation of the patient state that addresses some of these issues; and (3) demonstrate in a novel but representative clinical dataset that our alternative representation yields consistently better results for achieving optimal control, as measured by off-policy policy evaluation, compared to representations that do not incorporate missingness information.