AlphaZero and MuZero have achieved state-of-the-art (SOTA) performance in a wide range of domains, including board games and robotics, with discrete and continuous action spaces. However, to obtain an improved policy, they often require an excessively large number of simulations, especially for domains with large action spaces. As the simulation budget decreases, their performance drops significantly. In addition, many important real-world applications have combinatorial (or exponential) action spaces, making it infeasible to search directly over all possible actions. In this paper, we extend AlphaZero and MuZero to learn and plan in more complex multiagent (MA) Markov decision processes, where the action spaces increase exponentially with the number of agents. Our new algorithms, MA Gumbel AlphaZero and MA Gumbel MuZero, respectively without and with model learning, achieve superior performance on cooperative multiagent control problems, while reducing the number of environmental interactions by up to an order of magnitude compared to model-free approaches. In particular, we significantly improve prior performance when planning with much fewer simulation budgets. The code and appendix are available at https://github.com/tjuHaoXiaotian/MA-MuZero.

Despite many breakthroughs in recent years, it is still hard for MultiAgent Reinforcement Learning (MARL) algorithms to directly solve complex tasks in MultiAgent Systems (MASs) from scratch. In this work, we study how to use Automatic Curriculum Learning (ACL) to reduce the number of environmental interactions required to learn a good policy. In order to solve a difficult task, ACL methods automatically select a sequence of tasks (i.e., curricula). The idea is to obtain maximum learning progress towards the final task by continuously learning on tasks that match the current capabilities of the learners. The key question is how to measure the learning progress of the learner for better curriculum selection. We propose a novel ACL framework, PrOgRessive mulTiagent Automatic curricuLum (PORTAL), for MASs. PORTAL selects curricula according to two critera: 1) How difficult is a task, relative to the learners’ current abilities? 2) How similar is a task, relative to the final task? By learning a shared feature space between tasks, PORTAL is able to characterize different tasks based on the distribution of features and select those that are similar to the final task. Also, the shared feature space can effectively facilitate the policy transfer between curricula. Experimental results show that PORTAL can train agents to master extremely hard cooperative tasks, which can not be achieved with previous state-of-the-art MARL algorithms.

Designing novel biological sequences with desired properties is a significant challenge in biological science because of the extra large search space. The traditional design process usually involves multiple rounds of costly wet lab evaluations. To reduce the need for expensive wet lab experiments, machine learning methods are used to aid in designing biological sequences. However, the limited availability of biological sequences with known properties hinders the training of machine learning models, significantly restricting their applicability and performance. To fill this gap, we present ERLBioSeq, an Evolutionary Reinforcement Learning algorithm for BIOlogical SEQuence design. ERLBioSeq leverages the capability of reinforcement learning to learn without prior knowledge and the potential of evolutionary algorithms to enhance the exploration of reinforcement learning in the large search space of biological sequences. Additionally, to enhance the efficiency of biological sequence design, we developed a predictor for sequence screening in the biological sequence design process, which incorporates both the local and global sequence information. We evaluated the proposed method on three main types of biological sequence design tasks, including the design of DNA, RNA, and protein. The results demonstrate that the proposed method achieves significant improvement compared to the existing state-of-the-art methods.

Intermittent control problems are common in real world. The interactions between the decision maker and the executor can be discontinuous (intermittent) due to various types of interruptions, e.g. unstable communication channel. Due to intermittent interaction, agents are unable to acquire the state sent by the executor and cannot transmit actions to the executor within a period of time step, i.e. bidirectional blockage, which may lead to inefficiencies of reinforcement learning policies and prevent the executors from completing the task. Such problem is not well studied in the RL community. In this paper, we model Intermittent control problem as an Intermittent Control Markov Decision Process, i.e agents are expected to generate action sequences corresponding to the unavailable states and transmit them before disabling interactions to ensure the smooth and effective motion of executors. However, directly generating multiple future actions in the original action space has unnatural motion issue and exploration difficulty. We propose **M**ulti-step **A**ction **R**epre**S**entation (**MARS**), which encodes a sequence of actions from the original action space to a compact and decodable latent space. Then based on the latent action sequence representation, the mainstream RL methods can be easily optimized to learn a smooth and efficient motion policy. Extensive experiments on simulation tasks and real-world robotic grasping tasks show that MARS significantly improves the learning efficiency and final performances compared with existing baselines.

Hybrid action control tasks are common in the real world, which require controlling some discrete and continuous actions simultaneously. To solve these tasks, existing Deep Reinforcement learning (DRL) methods either directly build a separate policy for each type of action or simplify the hybrid action space into a discrete or continuous action control problem. However, these methods neglect the challenge of exploration resulting from the complexity of the hybrid action space. Thus, it is necessary to design more sample efficient algorithms. To this end, we propose a novel Hybrid Control Transformer (Hybrid CtrlFormer), to achieve better exploration and exploitation for the hybrid action control problems. The core idea is: 1) we construct a hybrid action space tree with the discrete actions at the higher level and the continuous parameter space at the lower level. Each parameter space is split into multiple subregions. 2) To simplify the exploration space, a Transformer-based Monte-Carlo tree search method is designed to efficiently evaluate and partition the hybrid action space into good and bad subregions along the tree. Our method achieves state-of-the-art performance and sample efficiency in a variety of environments with discrete-continuous action space.

In this paper, we focus on the decision-making challenge in autonomous driving, a central and intricate problem influencing the safety and practicality of autonomous vehicles. We propose an innovative hierarchical imitation learning framework that effectively alleviates the complexity of learning in autonomous driving decision-making problems by decoupling decision-making tasks into sub-problems. Specifically, the decision-making process is divided into two levels of sub-problems: the upper level directs the vehicle's lane selection and qualitative speed management, while the lower level implements precise control of the driving speed and direction. We harness Transformer-based models for solving each sub-problem, enabling overall hierarchical framework to comprehend and navigate diverse and various road conditions, ultimately resulting in improved decision-making. Through an evaluation in several typical driving scenarios within the SMARTS autonomous driving simulation environment, our proposed hierarchical decision-making framework significantly outperforms end-to-end reinforcement learning algorithms and behavior cloning algorithm, achieving an average pass rate of over 90%. Our framework's effectiveness is substantiated by its commendable achievements at the NeurIPS 2022 Driving SMARTS competition, where it secures dual track championships.

Multi-agent systems require effective coordination between groups and individuals to achieve common goals. However, current multi-agent reinforcement learning (MARL) methods primarily focus on improving individual policies and do not adequately address group-level policies, which leads to weak cooperation. To address this issue, we propose a novel Consensus-oriented Strategy (CoS) that emphasizes group and individual policies simultaneously. Specifically, CoS comprises two main components: (a) the vector quantized group consensus module, which extracts discrete latent embeddings that represent the stable and discriminative group consensus, and (b) the group consensus-oriented strategy, which integrates the group policy using a hypernet and the individual policies using the group consensus, thereby promoting coordination at both the group and individual levels. Through empirical experiments on cooperative navigation tasks with both discrete and continuous spaces, as well as google research football, we demonstrate that CoS outperforms state-of-the-art MARL algorithms and achieves better collaboration, thus providing a promising solution for achieving effective coordination in multi-agent systems.

Learning to collaborate is critical in Multi-Agent Reinforcement Learning (MARL). Previous works promote collaboration by maximizing the correlation of agents’ behaviors, which is typically characterized by Mutual Information (MI) in different forms. However, we reveal sub-optimal collaborative behaviors also emerge with strong correlations, and simply maximizing the MI can, surprisingly, hinder the learning towards better collaboration. To address this issue, we propose a novel MARL framework, called Progressive Mutual Information Collaboration (PMIC), for more effective MI-driven collaboration. PMIC uses a new collaboration criterion measured by the MI between global states and joint actions. Based on this criterion, the key idea of PMIC is maximizing the MI associated with superior collaborative behaviors and minimizing the MI associated with inferior ones. The two MI objectives play complementary roles by facilitating better collaborations while avoiding falling into sub-optimal ones. Experiments on a wide range of MARL benchmarks show the superior performance of PMIC compared with other algorithms.

Learning in real-world multiagent tasks is challenging due to the usual partial observability of each agent. Previous efforts alleviate the partial observability by historical hidden states with Recurrent Neural Networks, however, they do not consider the multiagent characters that either the multiagent observation consists of a number of object entities or the action space shows clear entity interactions. To tackle these issues, we propose the Agent Transformer Memory (ATM) network with a transformer-based memory. First, ATM utilizes the transformer to enable the unified processing of the factored environmental entities and memory. Inspired by the human’s working memory process where a limited capacity of information temporarily held in mind can effectively guide the decision-making, ATM updates its fixed-capacity memory with the working memory updating schema. Second, as agents' each action has its particular interaction entities in the environment, ATM parses the action space to introduce this action’s semantic inductive bias by binding each action with its specified involving entity to predict the state-action value or logit. Extensive experiments on the challenging SMAC and Level-Based Foraging environments validate that ATM could boost existing multiagent RL algorithms with impressive learning acceleration and performance improvement.

Given a data matrix \( D \) , a submatrix \( S \) of \( D \) is an order-preserving submatrix (OPSM) if there is a permutation of the columns of \( S \) , under which the entry values of each row in \( S \) are strictly increasing. OPSM mining is widely used in real-life applications such as identifying coexpressed genes and finding customers with similar preference. However, noise is ubiquitous in real data matrices due to variable experimental conditions and measurement errors, which makes conventional OPSM mining algorithms inapplicable. No previous work on OPSM has ever considered uncertain value intervals using the well-established possible world semantics. We establish two different definitions of significant OPSMs based on thepossible world semantics: (1) expected support-based and (2) probabilistic frequentness-based. An optimized dynamic programming approach is proposed to compute the probability that a row supports a particular column permutation, with a closed-form formula derived to efficiently handle the special case of uniform value distribution and an accurate cubic spline approximation approach that works well with any uncertain value distributions. To efficiently check the probabilistic frequentness, several effective pruning rules are designed to efficiently prune insignificant OPSMs; two approximation techniques based on the Poisson and Gaussian distributions, respectively, are proposed for further speedup. These techniques are integrated into our two OPSM mining algorithms, based on prefix-projection and Apriori, respectively. We further parallelize our prefix-projection-based mining algorithm using PrefixFPM, a recently proposed framework for parallel frequent pattern mining, and we achieve a good speedup with the number of CPU cores. Extensive experiments on real microarray data demonstrate that the OPSMs found by our algorithms have a much higher quality than those found by existing approaches.

The Dynamic Pickup and Delivery Problem (DPDP) is an essential problem in the logistics domain, which is NP-hard. The objective is to dynamically schedule vehicles among multiple sites to serve the online generated orders such that the overall transportation cost could be minimized. The critical challenge of DPDP is the orders are not known a priori, i.e., the orders are dynamically generated in real-time. To address this problem, existing methods partition the overall DPDP into fixed-size sub-problems by caching online generated orders and solve each sub-problem, or on this basis to utilize the predicted future orders to optimize each sub-problem further. However, the solution quality and efficiency of these methods are unsatisfactory, especially when the problem scale is very large. In this paper, we propose a novel hierarchical optimization framework to better solve large-scale DPDPs. Specifically, we design an upper-level agent to dynamically partition the DPDP into a series of sub-problems with different scales to optimize vehicles routes towards globally better solutions. Besides, a lower-level agent is designed to efficiently solve each sub-problem by incorporating the strengths of classical operational research-based methods with reinforcement learning-based policies. To verify the effectiveness of the proposed framework, real historical data is collected from the order dispatching system of Huawei Supply Chain Business Unit and used to build a functional simulator. Extensive offline simulation and online testing conducted on the industrial order dispatching system justify the superior performance of our framework over existing baselines.

A lot of efforts have been devoted to investigating how agents can learn effectively and achieve coordination in multiagent systems. However, it is still challenging in large-scale multiagent settings due to the complex dynamics between the environment and agents and the explosion of state-action space. In this paper, we design a novel Dynamic Multiagent Curriculum Learning (DyMA-CL) to solve large-scale problems by starting from learning on a multiagent scenario with a small size and progressively increasing the number of agents. We propose three transfer mechanisms across curricula to accelerate the learning process. Moreover, due to the fact that the state dimension varies across curricula, and existing network structures cannot be applied in such a transfer setting since their network input sizes are fixed. Therefore, we design a novel network structure called Dynamic Agent-number Network (DyAN) to handle the dynamic size of the network input. Experimental results show that DyMA-CL using DyAN greatly improves the performance of large-scale multiagent learning compared with state-of-the-art deep reinforcement learning approaches. We also investigate the influence of three transfer mechanisms across curricula through extensive simulations.

War-game is a type of multi-agent real-time strategy game, with challenges of the large-scale decision-making space and the flexible and changeable battlefield situation. In addition to the military field, it has played a role in fields including epidemic prevention and pest control. In recent years, more and more learning algorithms have tried to solve this kind of game. However, the existing methods have not yet given a satisfactory solution for the war-game, especially when preparation time is limited. In this background, we try to solve a traditional war-game based on hexagon grids. We propose a hierarchical multi-agent reinforcement learning framework to rapidly training an AI model for the war-game. The higher-level network in our hierarchical framework is used for task decision, it solves the credit assignment problem between agents through cooperative training. The lower-level network is mainly used for route planning, and it can be reused through parameter sharing for all the agents and all the maps. To deal with various opponents, we improve the robustness of the model through a grouped self-play approach. In experiments, we get encouraging results which show that the hierarchical structure allows agents to learn their strategies effectively. Our final AI model demonstrates that our methods can effectively deal with the challenges in the war-game.

In E-commerce, advertising is essential for merchants to reach their target users. The typical objective is to maximize the advertiser’s cumulative revenue over a period of time under a budget constraint. In real applications, an advertisement (ad) usually needs to be exposed to the same user multiple times until the user finally contributes revenue (e.g., places an order). However, existing advertising systems mainly focus on the immediate revenue with single ad exposures, ignoring the contribution of each exposure to the final conversion, thus usually falls into suboptimal solutions. In this paper, we formulate the sequential advertising strategy optimization as a dynamic knapsack problem. We propose a theoretically guaranteed bilevel optimization framework, which significantly reduces the solution space of the original optimization space while ensuring the solution quality. To improve the exploration efficiency of reinforcement learning, we also devise an effective action space reduction approach. Extensive offline and online experiments show the superior performance of our approaches over state-of-the-art baselines in terms of cumulative revenue.

Bipartite b-matching is fundamental in algorithm design, and has been widely applied into diverse applications, such as economic markets, labor markets, etc. These practical problems usually exhibit two distinct features: large-scale and dynamic, which requires the matching algorithm to be repeatedly executed at regular intervals. However, existing exact and approximate algorithms usually fail in such settings due to either requiring intolerable running time or too much computation resource. To address this issue, based on a key observation that the matching instances vary not too much, we propose NeuSearcher which leverage the knowledge learned from previously instances to solve new problem instances. Specifically, we design a multichannel graph neural network to predict the threshold of the matched edges, by which the search region could be significantly reduced. We further propose a parallel heuristic search algorithm to iteratively improve the solution quality until convergence. Experiments on both open and industrial datasets demonstrate that NeuSearcher can speed up 2 to 3 times while achieving exactly the same matching solution compared with the state-of-the-art approximation approaches.

BACKGROUND: Protein-protein interactions (PPIs) are of great importance in cellular systems of organisms, since they are the basis of cellular structure and function and many essential cellular processes are related to that. Most proteins perform their functions by interacting with other proteins, so predicting PPIs accurately is crucial for understanding cell physiology. RESULTS: Recently, graph convolutional networks (GCNs) have been proposed to capture the graph structure information and generate representations for nodes in the graph. In our paper, we use GCNs to learn the position information of proteins in the PPIs networks graph, which can reflect the properties of proteins to some extent. Combining amino acid sequence information and position information makes a stronger representation for protein, which improves the accuracy of PPIs prediction. CONCLUSION: In previous research methods, most of them only used protein amino acid sequence as input information to make predictions, without considering the structural information of PPIs networks graph. We first time combine amino acid sequence information and position information to make representations for proteins. The experimental results indicate that our method has strong competitiveness compared with several sequence-based methods.

Identification of protein-protein interactions (PPIs) is an important problem in biology, since PPIs are related to many essential cellular processes. The development of large-scale high-throughput experiments has produced a large number of PPIs data, however, these data are often noisy and their coverage is still limited. To overcome the shortcomings of experimental methods, many computational methods have been proposed for the prediction of PPIs. Among these methods, most of them solely take the amino acid sequence of protein as input information to make predictions. As PPIs data form the PPIs networks graph, the position information of proteins in the graph can reflect the properties of proteins to some extent, which is an important complement to protein sequence information. But previous works did not consider the graph structure information to improve the prediction performance. In this work, we first time apply graph convolutional networks (GCNs) to capture the protein's position information in the graph and combine amino acid sequence information and position information to make representations in the prediction task. Our experimental results show that our work outperforms the state-of-the-art sequence-based methods on several benchmark datasets and our work computationally is more efficient compared with previous works.

Precipitation nowcasting, which predicts rainfall intensity in the near future, has been studied by meteorologists for decades. Currently, computer vision techniques, especially optical flow based methods, are widely adopted by observatories since they deliver reasonable performance without the need of model training. However, their performance is highly sensitive to model parameters which require a lot of empirical knowledge to optimize. With the recent success of deep learning (DL), machine learning researchers have started to explore the use of spatiotemporal DL models for precipitation nowcasting, which have demonstrated a better performance than optical flow based methods. However, DL models are not easy to conFigure for nonDL experts such as meteorologists. In this poster, we introduce EasyRain, a platform with a user-friendly web interface to help users without domain knowledge (in DL and/or meteorology) to efficiently build DL and optical flow based models. We will demonstrate the efficiency and usability of EasyRain for training, tuning, and comparing precipitation nowcasting models.

In E-commerce advertising, where product recommendations and product ads are presented to users simultaneously, the traditional setting is to display ads at fixed positions. However, under such a setting, the advertising system loses the flexibility to control the number and positions of ads, resulting in sub-optimal platform revenue and user experience. Consequently, major e-commerce platforms (e.g., Taobao.com) have begun to consider more flexible ways to display ads. In this paper, we investigate the problem of advertising with adaptive exposure: can we dynamically determine the number and positions of ads for each user visit under certain business constraints so that the platform revenue can be increased? More specifically, we consider two types of constraints: request-level constraint ensures user experience for each user visit, and platform-level constraint controls the overall platform monetization rate. We model this problem as a Constrained Markov Decision Process with per-state constraint (psCMDP) and propose a constrained two-level reinforcement learning approach to decompose the original problem into two relatively independent sub-problems. To accelerate policy learning, we also devise a constrained hindsight experience replay mechanism. Experimental evaluations on industry-scale real-world datasets demonstrate the merits of our approach in both obtaining higher revenue under the constraints and the effectiveness of the constrained hindsight experience replay mechanism.

Given a data matrix D, a submatrix S of D is an order-preserving submatrix (OPSM) if there is a permutation of the columns of S, under which the entry values of each row in S are strictly increasing. OPSM mining is widely used in real-life applications such as identifying coexpressed genes, and finding customers with similar preference. However, noise is ubiquitous in real data matrices due to variable experimental conditions and measurement errors, which makes conventional OPSM mining algorithms inapplicable. No previous work has ever combated uncertain value intervals using the possible world semantics. We establish two different definitions of significant OPSMs based on the possible world semantics: (1) expected support based and (2) probabilistic frequentness based. An optimized dynamic programming approach is proposed to compute the probability that a row supports a particular column permutation, and several effective pruning rules are introduced to efficiently prune insignificant OPSMs. These techniques are integrated into our two OPSM mining algorithms, based on prefix-projection and Apriori respectively. Extensive experiments on real microarray data demonstrate that the OPSMs found by our algorithms have a much higher quality than those found by existing approaches.

BACKGROUND: The cooperation of cells in biological systems is similar to that of agents in cooperative multi-agent systems. Research findings in multi-agent systems literature can provide valuable inspirations to biological research. The well-coordinated states in cell systems can be viewed as desirable social norms in cooperative multi-agent systems. One important research question is how a norm can rapidly emerge with limited communication resources. RESULTS: In this work, we propose a learning approach which can trade off the agents' performance of coordinating on a consistent norm and the communication cost involved. During the learning process, the agents can dynamically adjust their coordination set according to their own observations and pick out the most crucial agents to coordinate with. In this way, our method significantly reduces the coordination dependence among agents. CONCLUSION: The experiment results show that our method can efficiently facilitate the social norm emergence among agents, and also scale well to large-scale populations.

The cooperation of cells in biological systems is similar to that of agents in cooperative multi-agent systems. Research findings in multi-agent systems literature can provide valuable inspirations to biological research. The well-coordinated states in cell systems can be viewed as desirable social norms in cooperative multi-agent systems. One important research question is how a norm can rapidly emerge with limited communication resources. In this paper, we propose a learning approach which can trade off the agents' performance of coordinating on a consistent norm and the communication cost involved. During the learning process, the agents can dynamically adjust their coordination set according to their own observations and pick out the most crucial agents to coordinate with. In this way, our method significantly reduces the coordination dependence among agents. The experiment results show that our method can efficiently facilitate the social norm emergence among agents, and also scale well to large-scale populations.