Almost all multi-agent reinforcement learning algorithms without communication follow the principle of centralized training with decentralized execution. During centralized training, agents can be guided by the same signals, such as the global state. During decentralized execution, however, agents lack the shared signal. Inspired by viewpoint invariance and contrastive learning, we propose consensus learning for cooperative multi-agent reinforcement learning in this paper. Although based on local observations, different agents can infer the same consensus in discrete space. During decentralized execution, we feed the inferred consensus as an explicit input to the network of agents, thereby developing their spirit of cooperation. Our proposed method can be extended to various multi-agent reinforcement learning algorithms with small model changes. Moreover, we carry out them on some fully cooperative tasks and get convincing results.

Understanding the dynamics of a system is important in many scientific and engineering domains. This problem can be approached by learning state transition rules from observations using machine learning techniques. Such observed time-series data often consist of sequences of many continuous variables with noise and ambiguity, but we often need rules of dynamics that can be modeled with a few essential variables. In this work, we propose a method for extracting a small number of essential hidden variables from high-dimensional time-series data and for learning state transition rules between these hidden variables. The proposed method is based on the Restricted Boltzmann Machine (RBM), which treats observable data in the visible layer and latent features in the hidden layer. However, real-world data, such as video and audio, include both discrete and continuous variables, and these variables have temporal relationships. Therefore, we propose Recurrent Temporal GaussianBernoulli Restricted Boltzmann Machine (RTGB-RBM), which combines Gaussian-Bernoulli Restricted Boltzmann Machine (GB-RBM) to handle continuous visible variables, and Recurrent Temporal Restricted Boltzmann Machine (RT-RBM) to capture time dependence between discrete hidden variables. We also propose a rule-based method that extracts essential information as hidden variables and represents state transition rules in interpretable form. We conduct experiments on Bouncing Ball and Moving MNIST datasets to evaluate our proposed method. Experimental results show that our method can learn the dynamics of those physical systems as state transition rules between hidden variables and can predict unobserved future states from observed state transitions.

We study the learning dynamics of self-predictive learning for reinforcement learning, a family of algorithms that learn representations by minimizing the prediction error of their own future latent representations. Despite its recent empirical success, such algorithms have an apparent defect: trivial representations (such as constants) minimize the prediction error, yet it is obviously undesirable to converge to such solutions. Our central insight is that careful designs of the optimization dynamics are critical to learning meaningful representations. We identify that a faster paced optimization of the predictor and semi-gradient updates on the representation, are crucial to preventing the representation collapse. Then in an idealized setup, we show self-predictive learning dynamics carries out spectral decomposition on the state transition matrix, effectively capturing information of the transition dynamics. Building on the theoretical insights, we propose bidirectional self-predictive learning, a novel self-predictive algorithm that learns two representations simultaneously. We examine the robustness of our theoretical insights with a number of small-scale experiments and showcase the promise of the novel representation learning algorithm with large-scale experiments.

Moving agents perceive streams of information, typically a mix of RGB images, depth and inertial measurements. Probabilistic generative models [1] are a principled way to formalise the synthesis of this data, and from these models inference can be derived through Bayes' rule. We focus on exactly such inference and target the agent states and the scene map, a problem known as simultaneous localisation and mapping (SLAM). We treat it as a posterior approximation for a given state-space model, such that the combination is useful for model-based control: the posterior inference serves as a state estimator and the predictive state-space model as a simulator with which to plan ahead [2]. To pave the way towards decision making, we believe an inference method should have: a compatible predictive model for both RGB-D images and 6-DoF dynamics; principled state and map uncertainty; real-time performance on commodity hardware; state-of-the-art localisation accuracy. We motivate these requirements further in appendix J. Prominent methods like LSD-SLAM [3], ORB-SLAM [4], DSO [5] have propelled visual SLAM forward, with heavy focus on large-scale localisation. The core of modern large-scale SLAM is maximum a-posteriori (MAP) smoothing in a probabilistic factor graph [6, 7]. At present this demands sparsity assumptions for computational feasibility, which obstructs the tight integration of dense maps and rendering.

Valuing mineral assets is a challenging task that is highly dependent on the supply (geological) uncertainty surrounding resources and reserves, and the uncertainty of demand (commodity prices). In this work, a graph-based reasoning, modeling and solution approach is proposed to jointly address mineral asset valuation and mine plan scheduling and optimization under supply and demand uncertainty in the "mining complex" framework. Three graph-based solutions are proposed: (i) a neural branching policy that learns a block-sampling ore body representation, (ii) a guiding policy that learns to explore a heuristic selection tree, (iii) a hyper-heuristic that manages the value/supply chain optimization and dynamics modeled as a graph structure. Results on two large-scale industrial mining complexes show a reduction of up to three orders of magnitude in primal suboptimality, execution time, and number of iterations, and an increase of up to 40% in the mineral asset value.

Recently, model-based agents have achieved better performance than model-free ones using the same computational budget and training time in single-agent environments. However, due to the complexity of multi-agent systems, it is tough to learn the model of the environment. The significant compounding error may hinder the learning process when model-based methods are applied to multi-agent tasks. This paper proposes an implicit model-based multi-agent reinforcement learning method based on value decomposition methods. Under this method, agents can interact with the learned virtual environment and evaluate the current state value according to imagined future states in the latent space, making agents have the foresight. Our approach can be applied to any multi-agent value decomposition method. The experimental results show that our method improves the sample efficiency in different partially observable Markov decision process domains.

Recent work in sim2real has successfully enabled robots to act in physical environments by training in simulation with a diverse ''population'' of environments (i.e. domain randomization). In this work, we focus on enabling generalization in assistive tasks: tasks in which the robot is acting to assist a user (e.g. helping someone with motor impairments with bathing or with scratching an itch). Such tasks are particularly interesting relative to prior sim2real successes because the environment now contains a human who is also acting. This complicates the problem because the diversity of human users (instead of merely physical environment parameters) is more difficult to capture in a population, thus increasing the likelihood of encountering out-of-distribution (OOD) human policies at test time. We advocate that generalization to such OOD policies benefits from (1) learning a good latent representation for human policies that test-time humans can accurately be mapped to, and (2) making that representation adaptable with test-time interaction data, instead of relying on it to perfectly capture the space of human policies based on the simulated population only. We study how to best learn such a representation by evaluating on purposefully constructed OOD test policies. We find that sim2real methods that encode environment (or population) parameters and work well in tasks that robots do in isolation, do not work well in assistance. In assistance, it seems crucial to train the representation based on the history of interaction directly, because that is what the robot will have access to at test time. Further, training these representations to then predict human actions not only gives them better structure, but also enables them to be fine-tuned at test-time, when the robot observes the partner act. https://adaptive-caregiver.github.io.

In the last decade, advanced approaches to the analysis and exploitation of large amounts of heterogeneous data ("big data") have gained tremendous attention, particularly on the part of corporate decision-makers but also from academic researchers [1, 2, 3]. The term "advanced analytics" generally refers to various methods beyond traditional multivariate statistics, mainly from the field of machine learning (ML), that leverage big data to drive decisions and actions (e.g., in organizations) [4, 5, 1, 6]. While researchers started to emphasize the suitability of these approaches mostly for (i) the design of innovative artifacts (e.g., decision support or process automation systems) and (ii) the induction of knowledge from quantitative studies [7, 1, 8, 9], companies increasingly deploy analytics applications in order to exploit their promising business potential [4, 6]. Several research articles show that such applications--especially those driven by modern ML algorithms--may considerably improve efficiency and/or effectiveness in important business areas, such as predictive maintenance, financial fraud detection, capacity planning, and product recommendation [10, 11, 12, 13, 14, 15, 16, 17]. The average return on investment of modern data analytics applications in a business context is estimated at an almost inconceivable rate of 1,301% [18].

This work studies networked agents cooperating to track a dynamical state of nature under partial information. The proposed algorithm is a distributed Bayesian filtering algorithm for finite-state hidden Markov models (HMMs). It can be used for sequential state estimation tasks, as well as for modeling opinion formation over social networks under dynamic environments. We show that the disagreement with the optimal centralized solution is asymptotically bounded for the class of geometrically ergodic state transition models, which includes rapidly changing models. We also derive recursions for calculating the probability of error and establish convergence under Gaussian observation models. Simulations are provided to illustrate the theory and to compare against alternative approaches.

Amazon and other e-commerce sites must employ mechanisms to protect their millions of customers from fraud, such as unauthorized use of credit cards. One such mechanism is order fraud evaluation, where systems evaluate orders for fraud risk, and either "pass" the order, or take an action to mitigate high risk. Order fraud evaluation systems typically use binary classification models that distinguish fraudulent and legitimate orders, to assess risk and take action. We seek to devise a system that considers both financial losses of fraud and long-term customer satisfaction, which may be impaired when incorrect actions are applied to legitimate customers. We propose that taking actions to optimize long-term impact can be formulated as a Reinforcement Learning (RL) problem. Standard RL methods require online interaction with an environment to learn, but this is not desirable in high-stakes applications like order fraud evaluation. Offline RL algorithms learn from logged data collected from the environment, without the need for online interaction, making them suitable for our use case. We show that offline RL methods outperform traditional binary classification solutions in SimStore, a simplified e-commerce simulation that incorporates order fraud risk. We also propose a novel approach to training offline RL policies that adds a new loss term during training, to better align policy exploration with taking correct actions.