Reinforcement learning has proven to be a valuable tool in the creation of advanced AI and robotic systems, contributing to everything from game playing to robotics to foundation models. Through trial-and-error, these AI systems typically learn one, near-optimal behavior to solve their tasks. However, there are many use cases in which one would like to assert some level of control, preferably in real time, over how the task is solved. We refer to these modifications of a core task as styles. We combine universal value function approximators (UVFAs) with carefully selected training scenarios, learning algorithms, and data augmentation to create a framework for coaching agents that exhibit styles in complex domains. We demonstrate the framework's application in the AAA video games Horizon Forbidden West and Gran Turismo, and in an open-source humanoid test domain. Despite the different nature of the domains – car racing, stylized game combat, and humanoid walking – each agent shows strong coherence to the style requests while still satisfying the main task in its domain. Importantly, the techniques outlined in this paper allow an end user to choose the final behavior at run time, giving them flexible control over the final executed performance.

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. Code available at: https://github.com/Danielhp95/on-opponent-modelling-in-expert-iteration-code. Supplementary material available at https://arxiv.org/abs/2206.00113.

Natural language instruction following is paramount to enable collaboration between artificial agents and human beings. Natural language-conditioned reinforcement learning (RL) agents have shown how natural languages' properties, such as compositionality, can provide a strong inductive bias to learn complex policies. Previous architectures like HIGhER combine the benefit of language-conditioning with Hindsight Experience Replay (HER) to deal with sparse rewards environments. Yet, like HER, HIGhER relies on an oracle predicate function to provide a feedback signal highlighting which linguistic description is valid for which state. This reliance on an oracle limits its application. Additionally, HIGhER only leverages the linguistic information contained in successful RL trajectories, thus hurting its final performance and data-efficiency. Without early successful trajectories, HIGhER is no better than DQN upon which it is built. In this paper, we propose the Emergent Textual Hindsight Experience Replay (ETHER) agent, which builds on HIGhER and addresses both of its limitations by means of (i) a discriminative visual referential game, commonly studied in the subfield of Emergent Communication (EC), used here as an unsupervised auxiliary task and (ii) a semantic grounding scheme to align the emergent language with the natural language of the instruction-following benchmark. We show that the referential game's agents make an artificial language emerge that is aligned with the natural-like language used to describe goals in the BabyAI benchmark and that it is expressive enough so as to also describe unsuccessful RL trajectories and thus provide feedback to the RL agent to leverage the linguistic, structured information contained in all trajectories. Our work shows that EC is a viable unsupervised auxiliary task for RL and provides missing pieces to make HER more widely applicable.

Modern reinforcement learning systems produce many high-quality policies throughout the learning process. However, to choose which policy to actually deploy in the real world, they must be tested under an intractable number of environmental conditions. We introduce RPOSST, an algorithm to select a small set of test cases from a larger pool based on a relatively small number of sample evaluations. RPOSST treats the test case selection problem as a two-player game and optimizes a solution with provable k-of-N robustness, bounding the error relative to a test that used all the test cases in the pool. Empirical results demonstrate that RPOSST finds a small set of test cases that identify high quality policies in a toy one-shot game, poker datasets, and a high-fidelity racing simulator.

The notion of self-play, albeit often cited in multiagent reinforcement learning as a process by which to train agent policies from scratch, has received little efforts to be taxonomized within a formal model. We present a formalized framework, with clearly defined assumptions, which encapsulates the meaning of self-play as abstracted from various existing self-play algorithms. This framework is framed as an approximation to a theoretical solution concept for multiagent training. Through a novel qualitative visualization metric, on a simple environment, we show that different self-play algorithms generate different distributions of episode trajectories, leading to different explorations of the policy space by the learning agents. Quantitatively, on two environments, we analyze the learning dynamics of policies trained under different self-play algorithms captured under our framework and perform cross self-play performance comparisons. Our results indicate that, throughout training, various widely used self-play algorithms exhibit cyclic policy evolutions and that the choice of self-play algorithm significantly affects the final performance of trained agents.

Automated game balancing has often focused on single-agent scenarios. In this paper we present a tool for balancing multi-player games during game design. Our approach requires a designer to construct an intuitive graphical representation of their meta-game target, representing the relative scores that high-level strategies (or decks, or character types) should experience. This permits more sophisticated balance targets to be defined beyond a simple requirement of equal win chances. We then find a parameterization of the game that meets this target using simulation-based optimization to minimize the distance to the target graph. We show the capabilities of this tool on examples inheriting from Rock-Paper-Scissors, and on a more complex asymmetric fighting game.

Throughout scientific history, overarching theoretical frameworks have allowed researchers to grow beyond personal intuitions and culturally biased theories. They allow to verify and replicate existing findings, and to link disconnected results. The notion of self-play, albeit often cited in multiagent Reinforcement Learning, has never been grounded in a formal model. We present a formalized framework, with clearly defined assumptions, which encapsulates the meaning of self-play as abstracted from various existing self-play algorithms. This framework is framed as an approximation to a theoretical solution concept for multiagent training. On a simple environment, we qualitatively measure how well a subset of the captured self-play methods approximate this solution when paired with the famous PPO algorithm. The results indicate that throughout training the trained policies exhibit cyclic evolutions, showing that self-play research is still at an early stage.

Deep reinforcement learning has recently been widely applied in robotics to study tasks such as locomotion and grasping, but its application to social human-robot interaction (HRI) remains a challenge. In this paper, we present a deep learning scheme that acquires a prior model of robot approaching behavior in simulation and applies it to real-world interaction with a physical robot approaching groups of humans. The scheme, which we refer to as Staged Social Behavior Learning (SSBL), considers different stages of learning in social scenarios. We learn robot approaching behaviors towards small groups in simulation and evaluate the performance of the model using objective and subjective measures in a perceptual study and a HRI user study with human participants. Results show that our model generates more socially appropriate behavior compared to a state-of-the-art model.

Introduccion: La prevalencia de las ulceras por presion en nuestro pais se estima alrededor del 7,87% a nivel hospitalario, llegando al 18% en las unidades de cuidados intensivos, donde hay mayor riesgo de desarrollar UPP. Ob j etivo: Comparar la habilidad predictiva de dos escalas de valoracion del riesgo de desarrollar UPP (escala Norton-MI e Indice COMHON) en una unidad de cuidados intensivos. M etodologia: Se trata de un estudio descriptivo retrospectivo. Como poblacion de estudio se incluyeron a todos los pacientes mayores de 18 anos ingresados en la UCI del hospital Obispo Polanco de Teruel desde el 1 de enero de 2017 hasta el 1 de marzo de 2018 que no presentaron ulceras por presion al ingreso y que tuvieron una estancia superior a 72 horas. La variable dependiente fue la presencia o ausencia de ulcera por presion. Las variables independientes fueron la edad, el sexo, el tiempo de estancia, la procedencia, el grupo diagnostico, la media de las puntuaciones de ambas escalas, la categoria y localizacion de la ulcera, la escala Apache II, el valor de prealbumina y los dias hasta la aparicion de la ulcera. Resultados: La sensibilidad, especificidad, VPP y VPN para la escala Norton- MI fue de 91,66%, 63,13%, 25% y 98,26% respectivamente, mientras que para el Indice COMHON fue de 95,83%, 32,4%, 15,97% y 98,3%. Conclusion: La validez predictiva fue ligeramente superior cuando se utiliza la escala Norton respecto al Indice Comhon. Ambas escalas fueron utiles para catalogar correctamente a los pacientes con bajo riesgo; sin embargo, dadas sus bajas especificidades y VPP, valores de alto riesgo no implican el desarrollo de ulceras por presion.

Deep reinforcement learning has been widely applied in the field of robotics recently to study tasks like locomotion and grasping, but applying it to social robotics remains a challenge. In this paper, we present a deep learning scheme that acquires a prior model of robot behavior in a simulator as a first phase to be further refined through learning from subsequent real-world interactions involving physical robots. The scheme, which we refer to as Staged Social Behavior Learning (SSBL), considers different stages of learning in social scenarios. Based on this scheme, we implement robot approaching behaviors towards a small group generated from F-formation and evaluate the performance of different configurations using objective and subjective measures. We found that our model generates more socially-considerate behavior compared to a state-of-the-art model, i.e. social force model. We also suggest that SSBL could be applied to a wide class of social robotics applications.