The challenge in controlling stochastic systems in which random events can set the system on catastrophic trajectories is to develop a robust ability to respond to such events without significantly compromising the optimality of the baseline control policy. Drug resistance can emerge from random and variable mutations in targeted cell populations; in the absence of an appropriate dosing policy, emergent resistant subpopulations can proliferate and lead to treatment failure. Dynamic feedback dosage control holds promise in combatting this phenomenon, but cell population evolutionary dynamics are complex, stochastic, and often high-dimensional, posing significant challenges to system control. This paper presents CelluDose, a deep reinforcement learning closed-loop dynamic control prototype for automated precision drug dosing targeting stochastic and heterogeneous cell proliferation. Developing optimal dosing schedules for preventing therapy-induced drug resistance involves a tradeoff between the effective suppression of emergent resistant cell subpopulations and the use of conservative dosages and a preference for first-line drugs. CelluDose is trained on model simulations of cell population evolutionary dynamics that combine a system of stochastic differential equations and the additional occurrence of random perturbing events. Both the single-drug and combination therapy policies obtained in training exhibit a 100% success rate at suppressing simulated heterogeneous harmful cell growth and responding to diverse system fluctuations and perturbations within the alloted time and using conservative dosing. The policies obtained were found to be highly robust to model parameter changes and fluctuations not introduced during training.

Recently, reinforcement learning (RL) algorithms have demonstrated remarkable success in learning complicated behaviors from minimally processed input. However, most of this success is limited to simulation. While there are promising successes in applying RL algorithms directly on real systems, their performance on more complex systems remains bottle-necked by the relative data inefficiency of RL algorithms. Domain randomization is a promising direction of research that has demonstrated impressive results using RL algorithms to control real robots. At a high level, domain randomization works by training a policy on a distribution of environmental conditions in simulation. If the environments are diverse enough, then the policy trained on this distribution will plausibly generalize to the real world. A human-specified design choice in domain randomization is the form and parameters of the distribution of simulated environments. It is unclear how to the best pick the form and parameters of this distribution and prior work uses hand-tuned distributions. This extended abstract demonstrates that the choice of the distribution plays a major role in the performance of the trained policies in the real world and that the parameter of this distribution can be optimized to maximize the performance of the trained policies in the real world

Reinforcement learning algorithms rely on exploration to discover new behaviors, which is typically achieved by following a stochastic policy. In continuous control tasks, policies with a Gaussian distribution have been widely adopted. Gaussian exploration however does not result in smooth trajectories that generally correspond to safe and rewarding behaviors in practical tasks. In addition, Gaussian policies do not result in an effective exploration of an environment and become increasingly inefficient as the action rate increases. This contributes to a low sample efficiency often observed in learning continuous control tasks. We introduce a family of stationary autoregressive (AR) stochastic processes to facilitate exploration in continuous control domains. We show that proposed processes possess two desirable features: subsequent process observations are temporally coherent with continuously adjustable degree of coherence, and the process stationary distribution is standard normal. We derive an autoregressive policy (ARP) that implements such processes maintaining the standard agent-environment interface. We show how ARPs can be easily used with the existing off-the-shelf learning algorithms. Empirically we demonstrate that using ARPs results in improved exploration and sample efficiency in both simulated and real world domains, and, furthermore, provides smooth exploration trajectories that enable safe operation of robotic hardware.

Various implementations of Deep Reinforcement Learning (RL) demonstrated excellent performance on tasks that can be solved by trained policy, but they are not without drawbacks. Deep RL suffers from high sensitivity to noisy and missing input and adversarial attacks. To mitigate these deficiencies of deep RL solutions, we suggest involving spiking neural networks (SNNs). Previous work has shown that standard Neural Networks trained using supervised learning for image classification can be converted to SNNs with negligible deterioration in performance. In this paper, we convert Q-Learning ReLU-Networks (ReLU-N) trained using reinforcement learning into SNN. We provide a proof of concept for the conversion of ReLU-N to SNN demonstrating improved robustness to occlusion and better generalization than the original ReLU-N. Moreover, we show promising initial results with converting full-scale Deep Q-networks to SNNs, paving the way for future research.

Recently, there have been several high-profile achievements of agents learning to play games against humans and beat them. We consider an alternative approach that instead addresses game design for a better player experience by training human-like game agents. Specifically, we study the problem of training game agents in service of the development processes of the game developers that design, build, and operate modern games. We highlight some of the ways in which we think intelligent agents can assist game developers to understand their games, and even to build them. Our early results using the proposed agent framework mark a few steps toward addressing the unique challenges that game developers face.

We examine the problem of adversarial reinforcement learning for multi-agent domains including a rule-based agent. Rule-based algorithms are required in safety-critical applications for them to work properly in a wide range of situations. Hence, every effort is made to find failure scenarios during the development phase. However, as the software becomes complicated, finding failure cases becomes difficult. Especially in multi-agent domains, such as autonomous driving environments, it is much harder to find useful failure scenarios that help us improve the algorithm. We propose a method for efficiently finding failure scenarios; this method trains the adversarial agents using multi-agent reinforcement learning such that the tested rule-based agent fails. We demonstrate the effectiveness of our proposed method using a simple environment and autonomous driving simulator.

In autonomous embedded systems, it is often vital to reduce the amount of actions taken in the real world and energy required to learn a policy. Training reinforcement learning agents from high dimensional image representations can be very expensive and time consuming. Autoencoders are deep neural network used to compress high dimensional data such as pixelated images into small latent representations. This compression model is vital to efficiently learn policies, especially when learning on embedded systems. We have implemented this model on the NVIDIA Jetson TX2 embedded GPU, and evaluated the power consumption, throughput, and energy consumption of the autoencoders for various CPU/GPU core combinations, frequencies, and model parameters. Additionally, we have shown the reconstructions generated by the autoencoder to analyze the quality of the generated compressed representation and also the performance of the reinforcement learning agent. Finally, we have presented an assessment of the viability of training these models on embedded systems and their usefulness in developing autonomous policies. Using autoencoders, we were able to achieve 4-5 $\times$ improved performance compared to a baseline RL agent with a convolutional feature extractor, while using less than 2W of power.

We introduce a novel apprenticeship learning algorithm to learn an expert's underlying reward structure in off-policy model-free \emph{batch} settings. Unlike existing methods that require a dynamics model or additional data acquisition for on-policy evaluation, our algorithm requires only the batch data of observed expert behavior. Such settings are common in real-world tasks---health care, finance or industrial processes ---where accurate simulators do not exist or data acquisition is costly. To address challenges in batch settings, we introduce Deep Successor Feature Networks(DSFN) that estimate feature expectations in an off-policy setting and a transition-regularized imitation network that produces a near-expert initial policy and an efficient feature representation. Our algorithm achieves superior results in batch settings on both control benchmarks and a vital clinical task of sepsis management in the Intensive Care Unit.

Using reinforcement learning to learn control policies is a challenge when the task is complex with potentially long horizons. Ensuring adequate but safe exploration is also crucial for controlling physical systems. In this paper, we use temporal logic to facilitate specification and learning of complex tasks. We combine temporal logic with control Lyapunov functions to improve exploration. We incorporate control barrier functions to safeguard the exploration and deployment process. We develop a flexible and learnable system that allows users to specify task objectives and constraints in different forms and at various levels. The framework is also able to take advantage of known system dynamics and handle unknown environmental dynamics by integrating model-free learning with model-based planning.

Reinforcement learning algorithms can be used to optimally solve dynamic decision-making and control problems. With continuous-valued state and input variables, reinforcement learning algorithms must rely on function approximators to represent the value function and policy mappings. Commonly used numerical approximators, such as neural networks or basis function expansions, have two main drawbacks: they are black-box models offering no insight in the mappings learned, and they require significant trial and error tuning of their meta-parameters. In this paper, we propose a new approach to constructing smooth value functions by means of symbolic regression. We introduce three off-line methods for finding value functions based on a state transition model: symbolic value iteration, symbolic policy iteration, and a direct solution of the Bellman equation. The methods are illustrated on four nonlinear control problems: velocity control under friction, one-link and two-link pendulum swing-up, and magnetic manipulation. The results show that the value functions not only yield well-performing policies, but also are compact, human-readable and mathematically tractable. This makes them potentially suitable for further analysis of the closed-loop system. A comparison with alternative approaches using neural networks shows that our method constructs well-performing value functions with substantially fewer parameters.