Growing up in north London, the child of a Greek Cypriot father and a Chinese Singaporean mother, Hassabis was a child prodigy in chess from the age of 4. He began writing his own computer games at 8, created one of the first video games to use AI at 17, and founded his own video game company not long after graduating from Cambridge University at 20. So perhaps it makes sense that Hassabis's AI startup DeepMind, founded in 2010 and sold to Google just four years later, would achieve its first major successes with AI models that used deep reinforcement learning to rapidly master video games like Space Invaders and Q*bert without any knowledge of the actual rules. That was followed with AlphaGo, which learned the ancient strategy board game of Go and would in 2017 defeat the world's number one human player -- an event that did perhaps more than anything else to awaken the world to the rapid progress of AI. New models could dominate a variety of games even faster, reducing the time and human intervention needed to acquire mastery.

How would an artificial intelligence (AI) decide what to do? One common approach in AI research is called "reinforcement learning". Reinforcement learning gives the software a "reward" defined in some way, and lets the software figure out how to maximise the reward. This approach has produced some excellent results, such as building software agents that defeat humans at games like chess and Go, or creating new designs for nuclear fusion reactors. However, we might want to hold off on making reinforcement learning agents too flexible and effective.

How would an artificial intelligence (AI) decide what to do? One common approach in AI research is called "reinforcement learning". Reinforcement learning gives the software a "reward" defined in some way, and lets the software figure out how to maximise the reward. This approach has produced some excellent results, such as building software agents that defeat humans at games like chess and Go, or creating new designs for nuclear fusion reactors. However, we might want to hold off on making reinforcement learning agents too flexible and effective.

Growing up in north London, the child of a Greek Cypriot father and a Chinese Singaporean mother, Hassabis was a child prodigy in chess from the age of 4. He began writing his own computer games at 8, created one of the first video games to use AI at 17, and founded his own video game company not long after graduating from Cambridge University at 20. So perhaps it makes sense that Hassabis's AI startup DeepMind, founded in 2010 and sold to Google just four years later, would achieve its first major successes with AI models that used deep reinforcement learning to rapidly master video games like Space Invaders and Q*bert without any knowledge of the actual rules. That was followed with AlphaGo, which learned the ancient strategy board game of Go and would in 2017 defeat the world's number one human player -- an event that did perhaps more than anything else to awaken the world to the rapid progress of AI. New models could dominate a variety of games even faster, reducing the time and human intervention needed to acquire mastery.

In probably approximately correct (PAC) reinforcement learning (RL), an agent is required to identify an $\epsilon$-optimal policy with probability $1-\delta$. While minimax optimal algorithms exist for this problem, its instance-dependent complexity remains elusive in episodic Markov decision processes (MDPs). In this paper, we propose the first nearly matching (up to a horizon squared factor and logarithmic terms) upper and lower bounds on the sample complexity of PAC RL in deterministic episodic MDPs with finite state and action spaces. In particular, our bounds feature a new notion of sub-optimality gap for state-action pairs that we call the deterministic return gap. While our instance-dependent lower bound is written as a linear program, our algorithms are very simple and do not require solving such an optimization problem during learning. Their design and analyses employ novel ideas, including graph-theoretical concepts (minimum flows) and a new maximum-coverage exploration strategy.

Unmanned aerial vehicles (UAVs) are finding use in applications that place increasing emphasis on robustness to external disturbances including extreme wind. However, traditional multirotor UAV platforms do not directly sense wind; conventional flow sensors are too slow, insensitive, or bulky for widespread integration on UAVs. Instead, drones typically observe the effects of wind indirectly through accumulated errors in position or trajectory tracking. In this work, we integrate a novel flow sensor based on micro-electro-mechanical systems (MEMS) hot-wire technology developed in our prior work onto a multirotor UAV for wind estimation. These sensors are omnidirectional, lightweight, fast, and accurate. In order to achieve superior tracking performance in windy conditions, we train a `wind-aware' residual-based controller via reinforcement learning using simulated wind gusts and their aerodynamic effects on the drone. In extensive hardware experiments, we demonstrate the wind-aware controller outperforming two strong `wind-unaware' baseline controllers in challenging windy conditions. See: https://youtu.be/KWqkH9Z-338.

This paper studies the computational offloading of CNN inference in dynamic multi-access edge computing (MEC) networks. To address the uncertainties in communication time and Edge servers' available capacity, we use early-exit mechanism to terminate the computation earlier to meet the deadline of inference tasks. We design a reward function to trade off the communication, computation and inference accuracy, and formulate the offloading problem of CNN inference as a maximization problem with the goal of maximizing the average inference accuracy and throughput in long term. To solve the maximization problem, we propose a graph reinforcement learning-based early-exit mechanism (GRLE), which outperforms the state-of-the-art work, deep reinforcement learning-based online offloading (DROO) and its enhanced method, DROO with early-exit mechanism (DROOE), under different dynamic scenarios. The experimental results show that GRLE achieves the average accuracy up to 3.41x over graph reinforcement learning (GRL) and 1.45x over DROOE, which shows the advantages of GRLE for offloading decision-making in dynamic MEC.

The multi-vehicle pursuit (MVP), as a problem abstracted from various real-world scenarios, is becoming a hot research topic in Intelligent Transportation System (ITS). The combination of Artificial Intelligence (AI) and connected vehicles has greatly promoted the research development of MVP. However, existing works on MVP pay little attention to the importance of information exchange and cooperation among pursuing vehicles under the complex urban traffic environment. This paper proposed a graded-Q reinforcement learning with information-enhanced state encoder (GQRL-IESE) framework to address this hierarchical collaborative multi-vehicle pursuit (HCMVP) problem. In the GQRL-IESE, a cooperative graded Q scheme is proposed to facilitate the decision-making of pursuing vehicles to improve pursuing efficiency. Each pursuing vehicle further uses a deep Q network (DQN) to make decisions based on its encoded state. A coordinated Q optimizing network adjusts the individual decisions based on the current environment traffic information to obtain the global optimal action set. In addition, an information-enhanced state encoder is designed to extract critical information from multiple perspectives and uses the attention mechanism to assist each pursuing vehicle in effectively determining the target. Extensive experimental results based on SUMO indicate that the total timestep of the proposed GQRL-IESE is less than other methods on average by 47.64%, which demonstrates the excellent pursuing efficiency of the GQRL-IESE. Codes are outsourced in https://github.com/ANT-ITS/GQRL-IESE.

Traditional control methods of robotic peg-in-hole assembly rely on complex contact state analysis. Reinforcement learning (RL) is gradually becoming a preferred method of controlling robotic peg-in-hole assembly tasks. However, the training process of RL is quite time-consuming because RL methods are always globally connected, which means all state components are assumed to be the input of policies for all action components, thus increasing action space and state space to be explored. In this paper, we first define continuous space serialized Shapley value (CS3) and construct a connection graph to clarify the correlativity of action components on state components. Then we propose a local connection reinforcement learning (LCRL) method based on the connection graph, which eliminates the influence of irrelevant state components on the selection of action components. The simulation and experiment results demonstrate that the control strategy obtained through LCRL method improves the stability and rapidity of the control process. LCRL method will enhance the data-efficiency and increase the final reward of the training process.

Despite impressive successes, deep reinforcement learning (RL) systems still fall short of human performance on generalization to new tasks and environments that differ from their training. As a benchmark tailored for studying RL generalization, we introduce Avalon, a set of tasks in which embodied agents in highly diverse procedural 3D worlds must survive by navigating terrain, hunting or gathering food, and avoiding hazards. Avalon is unique among existing RL benchmarks in that the reward function, world dynamics, and action space are the same for every task, with tasks differentiated solely by altering the environment; its 20 tasks, ranging in complexity from eat and throw to hunt and navigate, each create worlds in which the agent must perform specific skills in order to survive. This setup enables investigations of generalization within tasks, between tasks, and to compositional tasks that require combining skills learned from previous tasks. Avalon includes a highly efficient simulator, a library of baselines, and a benchmark with scoring metrics evaluated against hundreds of hours of human performance, all of which are open-source and publicly available. We find that standard RL baselines make progress on most tasks but are still far from human performance, suggesting Avalon is challenging enough to advance the quest for generalizable RL.