Lewis signaling games are a class of simple communication games for simulating the emergence of language. In these games, two agents must agree on a communication protocol in order to solve a cooperative task. Previous work has shown that agents trained to play this game with reinforcement learning tend to develop languages that display undesirable properties from a linguistic point of view (lack of generalization, lack of compositionality, etc). In this paper, we aim to provide better understanding of this phenomenon by analytically studying the learning problem in Lewis games. As a core contribution, we demonstrate that the standard objective in Lewis games can be decomposed in two components: a co-adaptation loss and an information loss. This decomposition enables us to surface two potential sources of overfitting, which we show may undermine the emergence of a structured communication protocol. In particular, when we control for overfitting on the co-adaptation loss, we recover desired properties in the emergent languages: they are more compositional and generalize better.

The navigation of robots in dynamic urban environments, requires elaborated anticipative strategies for the robot to avoid collisions with dynamic objects, like bicycles or pedestrians, and to be human aware. We have developed and analyzed three anticipative strategies in motion planning taking into account the future motion of the mobile objects that can move up to 18 km/h. First, we have used our hybrid policy resulting from a Deep Deterministic Policy Gradient (DDPG) training and the Social Force Model (SFM), and we have tested it in simulation in four complex map scenarios with many pedestrians. Second, we have used these anticipative strategies in real-life experiments using the hybrid motion planning method and the ROS Navigation Stack with Dynamic Windows Approach (NS-DWA). The results in simulations and real-life experiments show very good results in open environments and also in mixed scenarios with narrow spaces.

The main challenge of multiagent reinforcement learning is the difficulty of learning useful policies in the presence of other simultaneously learning agents whose changing behaviors jointly affect the environment's transition and reward dynamics. An effective approach that has recently emerged for addressing this non-stationarity is for each agent to anticipate the learning of other agents and influence the evolution of future policies towards desirable behavior for its own benefit. Unfortunately, previous approaches for achieving this suffer from myopic evaluation, considering only a finite number of policy updates. As such, these methods can only influence transient future policies rather than achieving the promise of scalable equilibrium selection approaches that influence the behavior at convergence. In this paper, we propose a principled framework for considering the limiting policies of other agents as time approaches infinity. Specifically, we develop a new optimization objective that maximizes each agent's average reward by directly accounting for the impact of its behavior on the limiting set of policies that other agents will converge to. Our paper characterizes desirable solution concepts within this problem setting and provides practical approaches for optimizing over possible outcomes. As a result of our farsighted objective, we demonstrate better long-term performance than state-of-the-art baselines across a suite of diverse multiagent benchmark domains.

We propose an extension to the MAPF formulation, called SocialMAPF, to account for private incentives of agents in constrained environments such as doorways, narrow hallways, and corridor intersections. SocialMAPF is able to, for instance, accurately reason about the urgent incentive of an agent rushing to the hospital over another agent's less urgent incentive of going to a grocery store; MAPF ignores such agent-specific incentives. Our proposed formulation addresses the open problem of optimal and efficient path planning for agents with private incentives. To solve SocialMAPF, we propose a new class of algorithms that use mechanism design during conflict resolution to simultaneously optimize agents' private local utilities and the global system objective. We perform an extensive array of experiments that show that optimal search-based MAPF techniques lead to collisions and increased time-to-goal in SocialMAPF compared to our proposed method using mechanism design. Furthermore, we empirically demonstrate that mechanism design results in models that maximizes agent utility and minimizes the overall time-to-goal of the entire system. We further showcase the capabilities of mechanism design-based planning by successfully deploying it in environments with static obstacles. To conclude, we briefly list several research directions using the SocialMAPF formulation, such as exploring motion planning in the continuous domain for agents with private incentives.

Is there a mathematical theory of intelligence? It all depends on how you define intelligence, as animal, human, machine or alien, or in the abstract terms, as a general mechanism transcending its specific operations and functions, as perception, cognitive processing, learning, reasoning, decision-making, and action. There are all sorts of models and approximations by means of mathematical logics, mathematical optimization, probability theory, statistic models, information theory, #computerscience. Accordingly, an intelligence could be realized in many forms and modalities, as an information entity, an animal/human being, an advanced algorithmic system, a learning software application, complex data-processing software/hardware, a sophisticated computing device, statistical machines, or a goal-directed agent, which "intelligence measures an agent's ability to achieve goals in a wide range of environments, situations, tasks and problems". The current situation is too divided, specific and fragmented.

We consider the general problem of geometric task allocation, wherein a large, decentralised swarm of simple mobile agents must detect the locations of tasks in the plane and position themselves nearby. The tasks are represented by an a priori unknown demand profile $\Phi(x,y)$ that determines how many agents are needed in each location. The agents are autonomous, oblivious and indistinguishable, and have finite sensing range. They must configure themselves according to $\Phi$ using only local information about $\Phi$ and about the positions of nearby agents. All agents act according to the same local sensing-based rule of motion, and cannot explicitly communicate nor share information. We propose an optimization-based approach to the problem which results in attraction-repulsion dynamics. Repulsion encourages agents to spread out and explore the region so as to find the tasks, and attraction causes them to accumulate at task locations. We derive this approach via gradient descent over an appropriate ``error'' functional, and test it extensively through numerical simulations. The figures in this work are snapshots of simulations that can be viewed online at https://youtu.be/kyUiGYSaaoQ.

Recent advances in deep reinforcement learning (RL) have demonstrated complex decision-making capabilities in simulation environments such as Arcade Learning Environment, MuJoCo, and ViZDoom. However, they are hardly extensible to more complicated problems, mainly due to the lack of complexity and variations in the environments they are trained and tested on. Furthermore, they are not extensible to an open-world environment to facilitate long-term exploration research. To learn realistic task-solving capabilities, we need to develop an environment with greater diversity and complexity. We developed WILD-SCAV, a powerful and extensible environment based on a 3D open-world FPS (First-Person Shooter) game to bridge the gap. It provides realistic 3D environments of variable complexity, various tasks, and multiple modes of interaction, where agents can learn to perceive 3D environments, navigate and plan, compete and cooperate in a human-like manner. WILD-SCAV also supports different complexities, such as configurable maps with different terrains, building structures and distributions, and multi-agent settings with cooperative and competitive tasks. The experimental results on configurable complexity, multi-tasking, and multi-agent scenarios demonstrate the effectiveness of WILD-SCAV in benchmarking various RL algorithms, as well as it is potential to give rise to intelligent agents with generalized task-solving abilities. The link to our open-sourced code can be found here https://github.com/inspirai/wilderness-scavenger.

There has been little exploration of the explicit simulation of the set of options of actors in agent-based models and its evolution over time. This study proposes to use affordances as intermediate entities between agents' environment and agent actions. We illustrated the approach on a typical gravity-fed network in the South-East of France to explore how the abandonment of traditional sharing of water changes the irrigators' options to irrigate. We simulated a typical dry year irrigation season under two institutional arrangements (i.e. traditional coordination through daily slots and its abandonment). Simulation results are consistent with field surveys, and reveal an increase in the number of internal conflicts among irrigators as the counterpart of the abandonment of traditional sharing of water. They also highlight the consequences of the heterogeneity of the irrigators' interests within the collective institution. The sensitivity analysis of the model allowed identification of optimal modalities of coordination, and a potential compromise between past and current institutional arrangements. The key benefits of using affordances in ABM lie in the study of their population dynamics for characterizing the interaction situations between actors and their environment and for better understanding the model dynamics.

Multi-agent reinforcement learning has drawn increasing attention in practice, e.g., robotics and automatic driving, as it can explore optimal policies using samples generated by interacting with the environment. However, high reward uncertainty still remains a problem when we want to train a satisfactory model, because obtaining high-quality reward feedback is usually expensive and even infeasible. To handle this issue, previous methods mainly focus on passive reward correction. At the same time, recent active reward estimation methods have proven to be a recipe for reducing the effect of reward uncertainty. In this paper, we propose a novel Distributional Reward Estimation framework for effective Multi-Agent Reinforcement Learning (DRE-MARL). Our main idea is to design the multi-action-branch reward estimation and policy-weighted reward aggregation for stabilized training. Specifically, we design the multi-action-branch reward estimation to model reward distributions on all action branches. Then we utilize reward aggregation to obtain stable updating signals during training. Our intuition is that consideration of all possible consequences of actions could be useful for learning policies. The superiority of the DRE-MARL is demonstrated using benchmark multi-agent scenarios, compared with the SOTA baselines in terms of both effectiveness and robustness.

Recent years have seen embodied visual navigation advance in two distinct directions: (i) in equipping the AI agent to follow natural language instructions, and (ii) in making the navigable world multimodal, e.g., audio-visual navigation. However, the real world is not only multimodal, but also often complex, and thus in spite of these advances, agents still need to understand the uncertainty in their actions and seek instructions to navigate. To this end, we present AVLEN~ -- an interactive agent for Audio-Visual-Language Embodied Navigation. Similar to audio-visual navigation tasks, the goal of our embodied agent is to localize an audio event via navigating the 3D visual world; however, the agent may also seek help from a human (oracle), where the assistance is provided in free-form natural language. To realize these abilities, AVLEN uses a multimodal hierarchical reinforcement learning backbone that learns: (a) high-level policies to choose either audio-cues for navigation or to query the oracle, and (b) lower-level policies to select navigation actions based on its audio-visual and language inputs. The policies are trained via rewarding for the success on the navigation task while minimizing the number of queries to the oracle. To empirically evaluate AVLEN, we present experiments on the SoundSpaces framework for semantic audio-visual navigation tasks. Our results show that equipping the agent to ask for help leads to a clear improvement in performance, especially in challenging cases, e.g., when the sound is unheard during training or in the presence of distractor sounds.