Researchers from the University of Illinois at Urbana-Champaign began with this more challenging task. They created a technique using multi-agent reinforcement learning (MARL), a form of artificial intelligence, to teach many agents to cooperate. Individual agents, such as robots or drones, can cooperate and finish a task when communication channels are open. What happens, though, if their technology is insufficient or the signals are jammed, making communication impossible? There are lots of research going on to improve the efficiency of artificial intelligence systems, lately, it is found that the selective regression method improves AI accuracy.

Now I am no historian, not by any stretch of the imagination; although, I always excelled at history in school. The history of Reinforcement Learning (RL) actually begins not in computer science, but in the field of psychology! If not, read on to learn the crude, but approximately optimal (RL insider joke) trajectory of RL history.

In recent years, both reinforcement learning and learning-based control -- as well as the study of their safety, which is crucial for deployment in real-world robots -- have gained significant traction. However, to adequately gauge the progress and applicability of new results, we need the tools to equitably compare the approaches proposed by the controls and reinforcement learning communities. Here, we propose a new open-source benchmark suite, called safe-control-gym, supporting both model-based and data-based control techniques. We provide implementations for three dynamic systems -- the cart-pole, the 1D, and 2D quadrotor -- and two control tasks -- stabilization and trajectory tracking. We propose to extend OpenAI's Gym API -- the de facto standard in reinforcement learning research -- with (i) the ability to specify (and query) symbolic dynamics and (ii) constraints, and (iii) (repeatably) inject simulated disturbances in the control inputs, state measurements, and inertial properties. To demonstrate our proposal and in an attempt to bring research communities closer together, we show how to use safe-control-gym to quantitatively compare the control performance, data efficiency, and safety of multiple approaches from the fields of traditional control, learning-based control, and reinforcement learning.

Deriving a good variable selection strategy in branch-and-bound is essential for the efficiency of modern mixed-integer programming (MIP) solvers. With MIP branching data collected during the previous solution process, learning to branch methods have recently become superior over heuristics. As branch-and-bound is naturally a sequential decision making task, one should learn to optimize the utility of the whole MIP solving process instead of being myopic on each step. In this work, we formulate learning to branch as an offline reinforcement learning (RL) problem, and propose a long-sighted hybrid search scheme to construct the offline MIP dataset, which values the long-term utilities of branching decisions. During the policy training phase, we deploy a ranking-based reward assignment scheme to distinguish the promising samples from the long-term or short-term view, and train the branching model named Branch Ranking via offline policy learning. Experiments on synthetic MIP benchmarks and real-world tasks demonstrate that Branch Rankink is more efficient and robust, and can better generalize to large scales of MIP instances compared to the widely used heuristics and state-of-the-art learning-based branching models.

The transformation towards intelligence in various industries is creating more demand for intelligent and flexible products. In the field of robotics, learning-based methods are increasingly being applied, with the purpose of training robots to learn to deal with complex and changing external environments through data. In this context, reinforcement learning and imitation learning are becoming research hotspots with their respective characteristics. However, the two have their own limitations in some cases, such as the high cost of data acquisition for reinforcement learning. Moreover, it is difficult for imitation learning to provide perfect demonstrations. As a branch of imitation learning, interactive imitation learning aims at transferring human knowledge to the agent through interactions between the demonstrator and the robot, which alleviates the difficulty of teaching. This thesis implements IIL algorithms in four simulation scenarios and conducts extensive experiments, aiming at providing exhaustive information about IIL methods both in action space and state space as well as comparison with RL methods.

Background and Motivation Industrial systems account for 54% of global energy usage [6] and 24% of global net anthropogenic Greenhouse Gas (GHG) emissions. The latter percentage rises to 34% if indirect emissions from energy are included, which would make industrial systems the highest emitting sector [35]. Due to increasing global demand for the products and services enabled by industrial systems, emissions from this sector will continue to rise [26]. However, there is strong evidence that interventions such as reduction in energy use per unit of output [38], lightweight designs and extended product lifetimes can facilitate critical emissions reductions across industrial systems [21]. Yet, optimizing industrial systems is not straightforward; subsectors such as metals, chemicals, waste and cement require customized approaches accounting for different materials, processes and facility configurations. Recent work has shown that reinforcement learning can be leveraged to efficiently control and optimize industrial processes (e.g.

One could imagine many simple heuristics, query scheduling with the explicit goal of reducing disk reads such as greedily selecting the next query with the highest and thus implicitly increasing query performance. We introduce expected buffer usage, to solve this problem. However, a SmartQueue, a learned scheduler that leverages overlapping hand-designed policy to handle the complexity of the entire data reads among incoming queries and learns a problem, including different buffer sizes, shifting query scheduling strategy that improves cache hits. SmartQueue workloads, heterogeneous data types (e.g., index files vs base relies on deep reinforcement learning to produce workloadspecific relations), and balancing short-term gains against long-term scheduling strategies that focus on long-term performance strategy is much more difficult to conceive.

Reinforcement learning (RL, see Sutton and Barto, 2018, for an overview) is concerned with how intelligence agents learn and take actions in an unknown environment in order to maximize the cumulative reward that it receives. It has been arguably one of the most vibrant research frontiers in machine learning over the last few years. According to Google Scholar, over 40K scientific articles have been published in 2020 with the phrase "reinforcement learning". Over 100 papers on RL were accepted for presentation at ICML 2021, a premier conference in the machine learning area, accounting for more than 10% of the accepted papers in total. RL algorithms have been applied in a wide variety of real applications, including games (Silver et al., 2016), robotics (Kormushev et al., 2013), healthcare (Komorowski et al., 2018), bidding (Jin et al., 2018), ridesharing (Xu et al., 2018) and automated driving (de Haan et al., 2019), to name a few. This paper is partly motivated by developing statistical learning methodologies in offline RL domains such as mobile health (mHealth).

Leveraging many sources of offline robot data requires grappling with the heterogeneity of such data. In this paper, we focus on one particular aspect of heterogeneity: learning from offline data collected at different control frequencies. Across labs, the discretization of controllers, sampling rates of sensors, and demands of a task of interest may differ, giving rise to a mixture of frequencies in an aggregated dataset. We study how well offline reinforcement learning (RL) algorithms can accommodate data with a mixture of frequencies during training. We observe that the $Q$-value propagates at different rates for different discretizations, leading to a number of learning challenges for off-the-shelf offline RL. We present a simple yet effective solution that enforces consistency in the rate of $Q$-value updates to stabilize learning. By scaling the value of $N$ in $N$-step returns with the discretization size, we effectively balance $Q$-value propagation, leading to more stable convergence. On three simulated robotic control problems, we empirically find that this simple approach outperforms na\"ive mixing by 50% on average.

Process synthesis experiences a disruptive transformation accelerated by digitization and artificial intelligence. We propose a reinforcement learning algorithm for chemical process design based on a state-of-the-art actor-critic logic. Our proposed algorithm represents chemical processes as graphs and uses graph convolutional neural networks to learn from process graphs. In particular, the graph neural networks are implemented within the agent architecture to process the states and make decisions. Moreover, we implement a hierarchical and hybrid decision-making process to generate flowsheets, where unit operations are placed iteratively as discrete decisions and corresponding design variables are selected as continuous decisions. We demonstrate the potential of our method to design economically viable flowsheets in an illustrative case study comprising equilibrium reactions, azeotropic separation, and recycles. The results show quick learning in discrete, continuous, and hybrid action spaces. Due to the flexible architecture of the proposed reinforcement learning agent, the method is predestined to include large action-state spaces and an interface to process simulators in future research.