This paper introduces the PettingZoo library and the accompanying Agent Environment Cycle ("AEC") games model. PettingZoo is a library of diverse sets of multi-agent environments with a universal, elegant Python API. PettingZoo was developed with the goal of accelerating research in Multi-Agent Reinforcement Learning ("MARL "), by making work more interchangeable, accessible and reproducible akin to what OpenAI's Gym library did for single-agent reinforcement

The training paradigm for large language models (LLMs) is moving from static datasets to experience-based learning, where agents acquire skills via interacting with complex environments. To facilitate this transition we introduce GEM (General Experience Maker), an open-source environment simulator designed for the age of LLMs. Analogous to OpenAI-Gym for traditional reinforcement learning (RL), GEM provides a standardized framework for the environment-agent interface, including asynchronous vectorized execution for high throughput, and flexible wrappers for easy extensibility. GEM also features a diverse suite of environments, robust integrated tools, and single-file example scripts demonstrating using GEM with five popular RL training frameworks. Along with this, we also provide a set of baselines across 24 environments using REINFORCE with Return Batch Normalization (ReBN), which -- unlike GRPO -- is compatible with the full RL setting of dense per-turn rewards and offers better credit assignment. We further conduct apple-to-apple benchmarking of PPO, GRPO and REINFORCE in both single- and multi-turn settings using GEM to shed light on the algorithmic designs. Lastly, GEM also functions as a convenient evaluation toolkit besides a training environment. We hope this framework can help accelerate future agentic LLM research.

This paper introduces the PettingZoo library and the accompanying Agent Environment Cycle ("AEC") games model. PettingZoo is a library of diverse sets of multi-agent environments with a universal, elegant Python API. PettingZoo was developed with the goal of accelerating research in Multi-Agent Reinforcement Learning ("MARL "), by making work more interchangeable, accessible and reproducible akin to what OpenAI's Gym library did for single-agent reinforcement

A huge event, with hundreds of participants, takeout pizza boxes stacked shoulder-high on carts, a jazz-rock band, a d.j., teams from about thirty high schools, robots by the dozen, and robot parts by the (probably) thousands spread out on tables in the cafeteria: it was the first day of the qualifiers for the all-city semifinals in the NYC FIRST Robotics Competition, at Francis Lewis High School, in Queens. On weekdays, about forty-four hundred students attend the school. In the rest of the building on this Saturday the hallways were empty. Michael Zigman, the C.E.O. of NYC FIRST, a nonprofit that provides STEM-education resources for students in public schools, stood in the gym, calculating in his head how many people were there. Zigman is a tall, kindly fifty-five-year-old Queens-born man who made money advising tech investors in the early two-thousands and then, in 2016, joined NYC FIRST.

This paper introduces the PettingZoo library and the accompanying Agent Environment Cycle ("AEC") games model. PettingZoo is a library of diverse sets of multi-agent environments with a universal, elegant Python API. PettingZoo was developed with the goal of accelerating research in Multi-Agent Reinforcement Learning ("MARL"), by making work more interchangeable, accessible and reproducible akin to what OpenAI's Gym library did for single-agent reinforcement learning. PettingZoo's API, while inheriting many features of Gym, is unique amongst MARL APIs in that it's based around the novel AEC games model. We argue, in part through case studies on major problems in popular MARL environments, that the popular game models are poor conceptual models of the games commonly used with MARL, that they promote severe bugs that are hard to detect, and that the AEC games model addresses these problems.

Animals can accomplish many incredible behavioral feats across a wide range of operational environments and scales that current robots struggle to match. One explanation for this performance gap is the extraordinary properties of the biological materials that comprise animals, such as muscle tissue. Using living muscle tissue as an actuator can endow robotic systems with highly desirable properties such as self-healing, compliance, and biocompatibility. Unlike traditional soft robotic actuators, living muscle biohybrid actuators exhibit unique adaptability, growing stronger with use. The dependency of a muscle's force output on its use history endows muscular organisms the ability to dynamically adapt to their environment, getting better at tasks over time. While muscle adaptability is a benefit to muscular organisms, it currently presents a challenge for biohybrid researchers: how does one design and control a robot whose actuators' force output changes over time? Here, we incorporate muscle adaptability into a many-muscle biohybrid robot design and modeling tool, leveraging reinforcement learning as both a co-design partner and system controller. As a controller, our learning agents coordinated the independent contraction of 42 muscles distributed on a lattice worm structure to successfully steer it towards eight distinct targets while incorporating muscle adaptability. As a co-design tool, our agents enable users to identify which muscles are important to accomplishing a given task. Our results show that adaptive agents outperform non-adaptive agents in terms of maximum rewards and training time. Together, these contributions can both enable the elucidation of muscle actuator adaptation and inform the design and modeling of adaptive, performant, many-muscle robots.

Chloe, an energetic young coach, promises to help you crush your fitness goals. The disciplined Rex, who has the air of a drill sergeant, encourages his clients to strive for excellence, but he is quick to warn that there won't be any shortcuts. If you're after a more mellow approach, Emma and Ethan are warm and quietly confident. But Lumin Fitness is no ordinary gym. These trainers don't exist--at least not physically.

They are also confident their system of AI trainers will encourage people to start working out even if they were previously put off gyms. The idea is to offer a more personalized approach to fitness that cuts out interactions with expert human trainers who could leave them feeling intimidated or unmotivated. The darkened studio space can accommodate up to 14 people at once, either completing a solo workout program or participating in a high-intensity functional training class where a group performs movements such as squats, dumbbell presses, and sit-ups. Each member works out within a designated station facing wall-to-wall LED screens. These tall screens mask sensors that track both the motions of the exerciser and the gym's specially built equipment, including dumbbells, medicine balls, and skipping ropes, using a combination of algorithms and machine-learning models.

Tracking climbers' activity to improve services and make the best use of their infrastructure is a concern for climbing gyms. Each climbing session must be analyzed from beginning till lowering of the climber. Therefore, spotting the climbers descending is crucial since it indicates when the ascent has come to an end. This problem must be addressed while preserving privacy and convenience of the climbers and the costs of the gyms. To this aim, a hardware prototype is developed to collect data using accelerometer sensors attached to a piece of climbing equipment mounted on the wall, called quickdraw, that connects the climbing rope to the bolt anchors. The corresponding sensors are configured to be energy-efficient, hence become practical in terms of expenses and time consumption for replacement when using in large quantity in a climbing gym. This paper describes hardware specifications, studies data measured by the sensors in ultra-low power mode, detect sensors' orientation patterns during lowering different routes, and develop an supervised approach to identify lowering.