The public goods game describes a social dilemma in which a large proportion of agents act as conditional cooperators (CC): they only act cooperatively if they see others acting cooperatively because they satisfice with the social norm to be in line with what others are doing instead of optimizing cooperation. CCs are guided by aspiration-based reinforcement learning guided by past experiences of interactions with others and satisficing aspirations. In many real-world settings, reinforcing social norms do not emerge. In this paper, we propose that an optimizing reinforcement agent can facilitate cooperation through nudges, i.e. indirect mechanisms for cooperation to happen. The agent's goal is to motivate CCs into cooperation through its own actions to create social norms that signal that others are cooperating. We introduce a multi-agent reinforcement learning model for public goods games, with 3 CC learning agents using aspirational reinforcement learning and 1 nudging agent using deep reinforcement learning to learn nudges that optimize cooperation. For our nudging agent, we model two distinct reward functions, one maximizing the total game return (sum DRL) and one maximizing the number of cooperative contributions contributions higher than a proportional threshold (prop DRL). Our results show that our aspiration-based RL model for CC agents is consistent with empirically observed CC behavior. Games combining 3 CC RL agents and one nudging RL agent outperform the baseline consisting of 4 CC RL agents only. The sum DRL nudging agent increases the total sum of contributions by 8.22% and the total proportion of cooperative contributions by 12.42%, while the prop nudging DRL increases the total sum of contributions by 8.85% and the total proportion of cooperative contributions by 14.87%. Our findings advance the literature on public goods games and reinforcement learning.

We develop a deep reinforcement learning method for training a jellyfish-like swimmer to effectively track a moving target in a two-dimensional flow. This swimmer is a flexible object equipped with a muscle model based on torsional springs. We employ a deep Q-network (DQN) that takes the swimmer's geometry and dynamic parameters as inputs, and outputs actions which are the forces applied to the swimmer. In particular, we introduce an action regulation to mitigate the interference from complex fluid-structure interactions. The goal of these actions is to navigate the swimmer to a target point in the shortest possible time. In the DQN training, the data on the swimmer's motions are obtained from simulations conducted using the immersed boundary method. During tracking a moving target, there is an inherent delay between the application of forces and the corresponding response of the swimmer's body due to hydrodynamic interactions between the shedding vortices and the swimmer's own locomotion. Our tests demonstrate that the swimmer, with the DQN agent and action regulation, is able to dynamically adjust its course based on its instantaneous state. This work extends the application scope of machine learning in controlling flexible objects within fluid environments.

In previous work, the IPMSRL environment (Integrated Platform Management System Reinforcement Learning environment) was developed with the aim of training defensive RL agents in a simulator representing a subset of an IPMS on a maritime vessel under a cyber-attack. This paper extends the use of IPMSRL to enhance realism including the additional dynamics of false positive alerts and alert delay. Applying curriculum learning, in the most difficult environment tested, resulted in an episode reward mean increasing from a baseline result of -2.791 to -0.569. Applying action masking, in the most difficult environment tested, resulted in an episode reward mean increasing from a baseline result of -2.791 to -0.743. Importantly, this level of performance was reached in less than 1 million timesteps, which was far more data efficient than vanilla PPO which reached a lower level of performance after 2.5 million timesteps. The training method which resulted in the highest level of performance observed in this paper was a combination of the application of curriculum learning and action masking, with a mean episode reward of 0.137. This paper also introduces a basic hardcoded defensive agent encoding a representation of cyber security best practice, which provides context to the episode reward mean figures reached by the RL agents. The hardcoded agent managed an episode reward mean of -1.895. This paper therefore shows that applications of curriculum learning and action masking, both independently and in tandem, present a way to overcome the complex real-world dynamics that are present in operational technology cyber security threat remediation.

Reusing pre-collected data from different domains is an attractive solution in decision-making tasks where the accessible data is insufficient in the target domain but relatively abundant in other related domains. Existing cross-domain policy transfer methods mostly aim at learning domain correspondences or corrections to facilitate policy learning, which requires learning domain/task-specific model components, representations, or policies that are inflexible or not fully reusable to accommodate arbitrary domains and tasks. These issues make us wonder: can we directly bridge the domain gap at the data (trajectory) level, instead of devising complicated, domain-specific policy transfer models? In this study, we propose a Cross-Domain Trajectory EDiting (xTED) framework with a new diffusion transformer model (Decision Diffusion Transformer, DDiT) that captures the trajectory distribution from the target dataset as a prior. The proposed diffusion transformer backbone captures the intricate dependencies among state, action, and reward sequences, as well as the transition dynamics within the target data trajectories. With the above pre-trained diffusion prior, source data trajectories with domain gaps can be transformed into edited trajectories that closely resemble the target data distribution through the diffusion-based editing process, which implicitly corrects the underlying domain gaps, enhancing the state realism and dynamics reliability in source trajectory data, while enabling flexible choices of downstream policy learning methods. Despite its simplicity, xTED demonstrates superior performance against other baselines in extensive simulation and real-robot experiments.

This report presents a solution for the swing-up and stabilisation tasks of the acrobot and the pendubot, developed for the AI Olympics competition at IROS 2024. Our approach employs the Average-Reward Entropy Advantage Policy Optimization (AR-EAPO), a model-free reinforcement learning (RL) algorithm that combines average-reward RL and maximum entropy RL. Results demonstrate that our controller achieves improved performance and robustness scores compared to established baseline methods in both the acrobot and pendubot scenarios, without the need for a heavily engineered reward function or system model. The current results are applicable exclusively to the simulation stage setup.

We have developed a reinforcement learning agent that often finds a minimal sequence of unknotting crossing changes for a knot diagram with up to 200 crossings, hence giving an upper bound on the unknotting number. We have used this to determine the unknotting number of 57k knots. We took diagrams of connected sums of such knots with oppositely signed signatures, where the summands were overlaid. The agent has found examples where several of the crossing changes in an unknotting collection of crossings result in hyperbolic knots. Based on this, we have shown that, given knots $K$ and $K'$ that satisfy some mild assumptions, there is a diagram of their connected sum and $u(K) + u(K')$ unknotting crossings such that changing any one of them results in a prime knot. As a by-product, we have obtained a dataset of 2.6 million distinct hard unknot diagrams; most of them under 35 crossings. Assuming the additivity of the unknotting number, we have determined the unknotting number of 43 at most 12-crossing knots for which the unknotting number is unknown.

Synthesizable molecular design (also known as synthesizable molecular optimization) is a fundamental problem in drug discovery, and involves designing novel molecular structures to improve their properties according to drug-relevant oracle functions (i.e., objective) while ensuring synthetic feasibility. However, existing methods are mostly based on random search. To address this issue, in this paper, we introduce a novel approach using the reinforcement learning method with quantum-inspired simulated annealing policy neural network to navigate the vast discrete space of chemical structures intelligently. Specifically, we employ a deterministic REINFORCE algorithm using policy neural networks to output transitional probability to guide state transitions and local search using genetic algorithm to refine solutions to a local optimum within each iteration. Our methods are evaluated with the Practical Molecular Optimization (PMO) benchmark framework with a 10K query budget. We further showcase the competitive performance of our method by comparing it against the state-of-the-art genetic algorithms-based method.

Abstract: Many approaches to robot learning begin by inferring a reward function from a set of human demonstrations. To learn a good reward, it is necessary to determine which features of the environment are relevant before determining how these features should be used to compute reward. End-to-end methods for joint feature and reward learning (e.g., using deep networks or program synthesis techniques) often yield brittle reward functions that are sensitive to spurious state features. By contrast, humans can often generalizably learn from a small number of demonstrations by incorporating strong priors about what features of a demonstration are likely meaningful for a task of interest. How do we build robots that leverage this kind of background knowledge when learning from new demonstrations? This paper describes a method named ALGAE (Adaptive Language-Guided Abstraction from [Contrastive] Explanations) which alternates between using language models to iteratively identify human-meaningful features needed to explain demonstrated behavior, then standard inverse reinforcement learning techniques to assign weights to these features. Experiments across a variety of both simulated and real-world robot environments show that ALGAE learns generalizable reward functions defined on interpretable features using only small numbers of demonstrations. Importantly, ALGAE can recognize when features are missing, then extract and define those features without any human input - making it possible to quickly and efficiently acquire rich representations of user behavior.

Recent successes in applying reinforcement learning (RL) for robotics has shown it is a viable approach for constructing robotic controllers. However, RL controllers can produce many collisions in environments where new obstacles appear during execution. This poses a problem in safety-critical settings. We present a hybrid approach, called iKinQP-RL, that uses an Inverse Kinematics Quadratic Programming (iKinQP) controller to correct actions proposed by an RL policy at runtime. This ensures safe execution in the presence of new obstacles not present during training. Preliminary experiments illustrate our iKinQP-RL framework completely eliminates collisions with new obstacles while maintaining a high task success rate.

In the domain of reinforcement learning, the formulation of a suitable reward function plays a pivotal role in shaping the behaviour of decision making agents. This is commonly justified by the widely adopted belief that the reward function is a succinct representation of a task goal in a given environment specified as a Markov decision process (MDP) (Ng et al., 2000). Eliciting the correct behavioural policies via the optimization of a reward function is of paramount importance for the deployment of RL agents to real world domains such as various robotics scenarios (Pomerleau, 1991; Billard et al., 2008) or expert behaviour forecasting (Kitani et al., 2012). However, the challenge of designing such a function typically entails a cumbersome and error-prone process of handcrafting a heuristic reward signal that accounts for all the intricacies of the task at hand. Inverse reinforcement learning (IRL) methods aim to solve the problem of inferring the reward function of an MDP based on a dataset of temporal behaviours.