This is technically Deep Learning in Python part 11 of my deep learning series, and my 3rd reinforcement learning course. Deep Reinforcement Learning is actually the combination of 2 topics: Reinforcement Learning and Deep Learning (Neural Networks). While both of these have been around for quite some time, it's only been recently that Deep Learning has really taken off, and along with it, Reinforcement Learning. The maturation of deep learning has propelled advances in reinforcement learning, which has been around since the 1980s, although some aspects of it, such as the Bellman equation, have been for much longer. Recently, these advances have allowed us to showcase just how powerful reinforcement learning can be.

We propose a general and systematic strategy to compile arbitrary quantum channels without using ancillary qubits, based on proximal policy optimization -- a powerful deep reinforcement learning algorithm. We rigorously prove that, in sharp contrast to the case of compiling unitary gates, it is impossible to compile an arbitrary channel to arbitrary precision with any given finite elementary channel set, regardless of the length of the decomposition sequence. However, for a fixed accuracy $\epsilon$ one can construct a universal set with constant number of $\epsilon$-dependent elementary channels, such that an arbitrary quantum channel can be decomposed into a sequence of these elementary channels followed by a unitary gate, with the sequence length bounded by $O(\frac{1}{\epsilon}\log\frac{1}{\epsilon})$. Through a concrete example concerning topological compiling of Majorana fermions, we show that our proposed algorithm can conveniently and effectively reduce the use of expensive elementary gates through adding the weighted cost into the reward function of the proximal policy optimization.

Defending computer networks from cyber attack requires timely responses to alerts and threat intelligence. Decisions about how to respond involve coordinating actions across multiple nodes based on imperfect indicators of compromise while minimizing disruptions to network operations. Currently, playbooks are used to automate portions of a response process, but often leave complex decision-making to a human analyst. In this work, we present a deep reinforcement learning approach to autonomous response and recovery in large industrial control networks. We propose an attention-based neural architecture that is flexible to the size of the network under protection. To train and evaluate the autonomous defender agent, we present an industrial control network simulation environment suitable for reinforcement learning. Experiments show that the learned agent can effectively mitigate advanced attacks that progress with few observable signals over several months before execution. The proposed deep reinforcement learning approach outperforms a fully automated playbook method in simulation, taking less disruptive actions while also defending more nodes on the network. The learned policy is also more robust to changes in attacker behavior than playbook approaches.

The optimistic nature of the Q-learning target leads to an overestimation bias, which is an inherent problem associated with standard $Q-$learning. Such a bias fails to account for the possibility of low returns, particularly in risky scenarios. However, the existence of biases, whether overestimation or underestimation, need not necessarily be undesirable. In this paper, we analytically examine the utility of biased learning, and show that specific types of biases may be preferable, depending on the scenario. Based on this finding, we design a novel reinforcement learning algorithm, Balanced Q-learning, in which the target is modified to be a convex combination of a pessimistic and an optimistic term, whose associated weights are determined online, analytically. We prove the convergence of this algorithm in a tabular setting, and empirically demonstrate its superior learning performance in various environments.

Reinforcement learning (RL) for continuous control typically employs distributions whose support covers the entire action space. In this work, we investigate the colloquially known phenomenon that trained agents often prefer actions at the boundaries of that space. We draw theoretical connections to the emergence of bang-bang behavior in optimal control, and provide extensive empirical evaluation across a variety of recent RL algorithms. We replace the normal Gaussian by a Bernoulli distribution that solely considers the extremes along each action dimension - a bang-bang controller. Surprisingly, this achieves state-of-the-art performance on several continuous control benchmarks - in contrast to robotic hardware, where energy and maintenance cost affect controller choices. Since exploration, learning,and the final solution are entangled in RL, we provide additional imitation learning experiments to reduce the impact of exploration on our analysis. Finally, we show that our observations generalize to environments that aim to model real-world challenges and evaluate factors to mitigate the emergence of bang-bang solutions. Our findings emphasize challenges for benchmarking continuous control algorithms, particularly in light of potential real-world applications.

Imitation learning (IL) is a popular approach in the continuous control setting as among other reasons it circumvents the problems of reward mis-specification and exploration in reinforcement learning (RL). In IL from demonstrations, an important challenge is to obtain agent policies that are smooth with respect to the inputs. Learning through imitation a policy that is smooth as a function of a large state-action ($s$-$a$) space (typical of high dimensional continuous control environments) can be challenging. We take a first step towards tackling this issue by using smoothness inducing regularizers on \textit{both} the policy and the cost models of adversarial imitation learning. Our regularizers work by ensuring that the cost function changes in a controlled manner as a function of $s$-$a$ space; and the agent policy is well behaved with respect to the state space. We call our new smooth IL algorithm \textit{Smooth Policy and Cost Imitation Learning} (SPaCIL, pronounced 'Special'). We introduce a novel metric to quantify the smoothness of the learned policies. We demonstrate SPaCIL's superior performance on continuous control tasks from MuJoCo. The algorithm not just outperforms the state-of-the-art IL algorithm on our proposed smoothness metric, but, enjoys added benefits of faster learning and substantially higher average return.

Reinforcement learning methods for continuous control tasks have evolved in recent years generating a family of policy gradient methods that rely primarily on a Gaussian distribution for modeling a stochastic policy. However, the Gaussian distribution has an infinite support, whereas real world applications usually have a bounded action space. This dissonance causes an estimation bias that can be eliminated if the Beta distribution is used for the policy instead, as it presents a finite support. In this work, we investigate how this Beta policy performs when it is trained by the Proximal Policy Optimization (PPO) algorithm on two continuous control tasks from OpenAI gym. For both tasks, the Beta policy is superior to the Gaussian policy in terms of agent's final expected reward, also showing more stability and faster convergence of the training process. For the CarRacing environment with high-dimensional image input, the agent's success rate was improved by 63% over the Gaussian policy.

Shared e-mobility services have been widely tested and piloted in cities across the globe, and already woven into the fabric of modern urban planning. This paper studies a practical yet important problem in those systems: how to deploy and manage their infrastructure across space and time, so that the services are ubiquitous to the users while sustainable in profitability. However, in real-world systems evaluating the performance of different deployment strategies and then finding the optimal plan is prohibitively expensive, as it is often infeasible to conduct many iterations of trial-and-error. We tackle this by designing a high-fidelity simulation environment, which abstracts the key operation details of the shared e-mobility systems at fine-granularity, and is calibrated using data collected from the real-world. This allows us to try out arbitrary deployment plans to learn the optimal given specific context, before actually implementing any in the real-world systems. In particular, we propose a novel multi-agent neural search approach, in which we design a hierarchical controller to produce tentative deployment plans. The generated deployment plans are then tested using a multi-simulation paradigm, i.e., evaluated in parallel, where the results are used to train the controller with deep reinforcement learning. With this closed loop, the controller can be steered to have higher probability of generating better deployment plans in future iterations. The proposed approach has been evaluated extensively in our simulation environment, and experimental results show that it outperforms baselines e.g., human knowledge, and state-of-the-art heuristic-based optimization approaches in both service coverage and net revenue. On the other hand, the recent work in [2] considers Shared electric mobility (e-mobility) systems are becoming the incremental cases, but essentially it uses greedy-based ubiquitous and forming a considerable part of our transportation approaches to re-compute for charger planning, which may paradigm in urban environments.

Artificial intelligence has proven that complicated board and video games are no longer the exclusive domain of the human mind. From chess to Go to StarCraft, AI systems that use reinforcement learning algorithms have outperformed human world champions in recent years. But despite the high individual performance of RL agents, they can become frustrating teammates when paired with human players, according to a study by AI researchers at MIT Lincoln Laboratory. The study, which involved cooperation between humans and AI agents in the card game Hanabi, shows that players prefer the classic and predictable rule-based AI systems over complex RL systems. The findings, presented in a paper published on arXiv, highlight some of the underexplored challenges of applying reinforcement learning to real-world situations and can have important implications for the future development of AI systems that are meant to cooperate with humans.

This article is part of our reviews of AI research papers, a series of posts that explore the latest findings in artificial intelligence. Artificial intelligence has proven that complicated board and video games are no longer the exclusive domain of the human mind. From chess to Go to StarCraft, AI systems that use reinforcement learning algorithms have outperformed human world champions in recent years. But despite the high individual performance of RL agents, they can become frustrating teammates when paired with human players, according to a study by AI researchers at MIT Lincoln Laboratory. The study, which involved cooperation between humans and AI agents in the card game Hanabi, shows that players prefer the classic and predictable rule-based AI systems over complex RL systems.