Traditional multi-agent reinforcement learning algorithms are difficultly applied in a large-scale multi-agent environment. The introduction of mean field theory has enhanced the scalability of multi-agent reinforcement learning in recent years. This paper considers partially observable multi-agent reinforcement learning (MARL), where each agent can only observe other agents within a fixed range. This partial observability affects the agent's ability to assess the quality of the actions of surrounding agents. This paper focuses on developing a method to capture more effective information from local observations in order to select more effective actions. Previous work in this field employs probability distributions or weighted mean field to update the average actions of neighborhood agents, but it does not fully consider the feature information of surrounding neighbors and leads to a local optimum. In this paper, we propose a novel multi-agent reinforcement learning algorithm, Partially Observable Mean Field Multi-Agent Reinforcement Learning based on Graph--Attention (GAMFQ) to remedy this flaw. GAMFQ uses a graph attention module and a mean field module to describe how an agent is influenced by the actions of other agents at each time step. This graph attention module consists of a graph attention encoder and a differentiable attention mechanism, and this mechanism outputs a dynamic graph to represent the effectiveness of neighborhood agents against central agents. The mean--field module approximates the effect of a neighborhood agent on a central agent as the average effect of effective neighborhood agents. We evaluate GAMFQ on three challenging tasks in the MAgents framework. Experiments show that GAMFQ outperforms baselines including the state-of-the-art partially observable mean-field reinforcement learning algorithms.

Finding a best response policy is a central objective in game theory and multi-agent learning, with modern population-based training approaches employing reinforcement learning algorithms as best-response oracles to improve play against candidate opponents (typically previously learnt policies). We propose Best Response Expert Iteration (BRExIt), which accelerates learning in games by incorporating opponent models into the state-of-the-art learning algorithm Expert Iteration (ExIt). BRExIt aims to (1) improve feature shaping in the apprentice, with a policy head predicting opponent policies as an auxiliary task, and (2) bias opponent moves in planning towards the given or learnt opponent model, to generate apprentice targets that better approximate a best response. In an empirical ablation on BRExIt's algorithmic variants against a set of fixed test agents, we provide statistical evidence that BRExIt learns better performing policies than ExIt.

Digital Twin (DT) is a prominent technology to utilise and deploy within the healthcare sector. Yet, the main challenges facing such applications are: Strict health data-sharing policies, high-performance network requirements, and possible infrastructure resource limitations. In this paper, we address all the challenges by provisioning adaptive Virtual Network Functions (VNFs) to enforce security policies associated with different data-sharing scenarios. We define a Cloud-Native Network orchestrator on top of a multi-node cluster mesh infrastructure for flexible and dynamic container scheduling. The proposed framework considers the intended data-sharing use case, the policies associated, and infrastructure configurations, then provision Service Function Chaining (SFC) and provides routing configurations accordingly with little to no human intervention. Moreover, what is \textit{optimal} when deploying SFC is dependent on the use case itself, and we tune the hyperparameters to prioritise resource utilisation or latency in an effort to comply with the performance requirements. As a result, we provide an adaptive network orchestration for digital health twin use cases, that is policy-aware, requirements-aware, and resource-aware.

Inverse reinforcement learning (IRL) denotes a powerful family of algorithms for recovering a reward function justifying the behavior demonstrated by an expert agent. A well-known limitation of IRL is the ambiguity in the choice of the reward function, due to the existence of multiple rewards that explain the observed behavior. This limitation has been recently circumvented by formulating IRL as the problem of estimating the feasible reward set, i.e., the region of the rewards compatible with the expert's behavior. In this paper, we make a step towards closing the theory gap of IRL in the case of finite-horizon problems with a generative model. We start by formally introducing the problem of estimating the feasible reward set, the corresponding PAC requirement, and discussing the properties of particular classes of rewards. Then, we provide the first minimax lower bound on the sample complexity for the problem of estimating the feasible reward set of order ${\Omega}\Bigl( \frac{H^3SA}{\epsilon^2} \bigl( \log \bigl(\frac{1}{\delta}\bigl) + S \bigl)\Bigl)$, being $S$ and $A$ the number of states and actions respectively, $H$ the horizon, $\epsilon$ the desired accuracy, and $\delta$ the confidence. We analyze the sample complexity of a uniform sampling strategy (US-IRL), proving a matching upper bound up to logarithmic factors. Finally, we outline several open questions in IRL and propose future research directions.

With increasing complexity of modern communication systems, machine learning algorithms have become a focal point of research. However, performance demands have tightened in parallel to complexity. For some of the key applications targeted by future wireless, such as the medical field, strict and reliable performance guarantees are essential, but vanilla machine learning methods have been shown to struggle with these types of requirements. Therefore, the question is raised whether these methods can be extended to better deal with the demands imposed by such applications. In this paper, we look at a combinatorial resource allocation challenge with rare, significant events which must be handled properly. We propose to treat this as a multi-task learning problem, select two methods from this domain, Elastic Weight Consolidation and Gradient Episodic Memory, and integrate them into a vanilla actor-critic scheduler. We compare their performance in dealing with Black Swan Events with the state-of-the-art of augmenting the training data distribution and report that the multi-task approach proves highly effective.

Inverted landing in a rapid and robust manner is a challenging feat for aerial robots, especially while depending entirely on onboard sensing and computation. In spite of this, this feat is routinely performed by biological fliers such as bats, flies, and bees. Our previous work has identified a direct causal connection between a series of onboard visual cues and kinematic actions that allow for reliable execution of this challenging aerobatic maneuver in small aerial robots. In this work, we first utilized Deep Reinforcement Learning and a physics-based simulation to obtain a general, optimal control policy for robust inverted landing starting from any arbitrary approach condition. This optimized control policy provides a computationally-efficient mapping from the system's observational space to its motor command action space, including both triggering and control of rotational maneuvers. This was done by training the system over a large range of approach flight velocities that varied with magnitude and direction. Next, we performed a sim-to-real transfer and experimental validation of the learned policy via domain randomization, by varying the robot's inertial parameters in the simulation. Through experimental trials, we identified several dominant factors which greatly improved landing robustness and the primary mechanisms that determined inverted landing success. We expect the learning framework developed in this study can be generalized to solve more challenging tasks, such as utilizing noisy onboard sensory data, landing on surfaces of various orientations, or landing on dynamically-moving surfaces.

This paper introduces two learning schemes for distributed agents in Reinforcement Learning (RL) environments, namely Reward-Weighted (R-Weighted) and Loss-Weighted (L-Weighted) gradient merger. The R/L weighted methods replace standard practices for training multiple agents, such as summing or averaging the gradients. The core of our methods is to scale the gradient of each actor based on how high the reward (for R-Weighted) or the loss (for L-Weighted) is compared to the other actors. During training, each agent operates in differently initialized versions of the same environment, which gives different gradients from different actors. In essence, the R-Weights and L-Weights of each agent inform the other agents of its potential, which again reports which environment should be prioritized for learning. This approach of distributed learning is possible because environments that yield higher rewards, or low losses, have more critical information than environments that yield lower rewards or higher losses. We empirically demonstrate that the R-Weighted methods work superior to the state-of-the-art in multiple RL environments.

Auxiliary tasks improve the representations learned by deep reinforcement learning agents. Analytically, their effect is reasonably well understood; in practice, however, their primary use remains in support of a main learning objective, rather than as a method for learning representations. This is perhaps surprising given that many auxiliary tasks are defined procedurally, and hence can be treated as an essentially infinite source of information about the environment. Based on this observation, we study the effectiveness of auxiliary tasks for learning rich representations, focusing on the setting where the number of tasks and the size of the agent's network are simultaneously increased. For this purpose, we derive a new family of auxiliary tasks based on the successor measure. These tasks are easy to implement and have appealing theoretical properties. Combined with a suitable off-policy learning rule, the result is a representation learning algorithm that can be understood as extending Mahadevan & Maggioni (2007)'s proto-value functions to deep reinforcement learning -- accordingly, we call the resulting object proto-value networks. Through a series of experiments on the Arcade Learning Environment, we demonstrate that proto-value networks produce rich features that may be used to obtain performance comparable to established algorithms, using only linear approximation and a small number (~4M) of interactions with the environment's reward function.

Machine learning (ML) based approaches are increasingly being used in a number of applications with societal impact. Training ML models often require vast amounts of labeled data, and crowdsourcing is a dominant paradigm for obtaining labels from multiple workers. Crowd workers may sometimes provide unreliable labels, and to address this, truth discovery (TD) algorithms such as majority voting are applied to determine the consensus labels from conflicting worker responses. However, it is important to note that these consensus labels may still be biased based on sensitive attributes such as gender, race, or political affiliation. Even when sensitive attributes are not involved, the labels can be biased due to different perspectives of subjective aspects such as toxicity. In this paper, we conduct a systematic study of the bias and fairness of TD algorithms. Our findings using two existing crowd-labeled datasets, reveal that a non-trivial proportion of workers provide biased results, and using simple approaches for TD is sub-optimal. Our study also demonstrates that popular TD algorithms are not a panacea. Additionally, we quantify the impact of these unfair workers on downstream ML tasks and show that conventional methods for achieving fairness and correcting label biases are ineffective in this setting. We end the paper with a plea for the design of novel bias-aware truth discovery algorithms that can ameliorate these issues.

Background: Computational drug repurposing is a cost- and time-efficient approach that aims to identify new therapeutic targets or diseases (indications) of existing drugs/compounds. It is especially critical for emerging and/or orphan diseases due to its cheaper investment and shorter research cycle compared with traditional wet-lab drug discovery approaches. However, the underlying mechanisms of action (MOAs) between repurposed drugs and their target diseases remain largely unknown, which is still a main obstacle for computational drug repurposing methods to be widely adopted in clinical settings. Results: In this work, we propose KGML-xDTD: a Knowledge Graph-based Machine Learning framework for explainably predicting Drugs Treating Diseases. It is a two-module framework that not only predicts the treatment probabilities between drugs/compounds and diseases but also biologically explains them via knowledge graph (KG) path-based, testable mechanisms of action (MOAs). We leverage knowledge-and-publication based information to extract biologically meaningful "demonstration paths" as the intermediate guidance in the Graph-based Reinforcement Learning (GRL) path-finding process. Comprehensive experiments and case study analyses show that the proposed framework can achieve state-of-the-art performance in both predictions of drug repurposing and recapitulation of human-curated drug MOA paths. Conclusions: KGML-xDTD is the first model framework that can offer KG-path explanations for drug repurposing predictions by leveraging the combination of prediction outcomes and existing biological knowledge and publications. We believe it can effectively reduce "black-box" concerns and increase prediction confidence for drug repurposing based on predicted path-based explanations, and further accelerate the process of drug discovery for emerging diseases.