FootsiesGym: A Fighting Game Benchmark for Two-Player Zero-Sum Imperfect-Information Games
Researchers released FootsiesGym, an open-source benchmark for training AI in two-player fighting games with imperfect information. The environment simplifies complex interactions while enabling high-throughput reinforcement learning.
- FootsiesGym is an open-source benchmark for two-player, zero-sum, imperfect-information games, built on a simplified fighting game.
- The environment includes a vectorized simulator for high-throughput training on standard hardware.
- Researchers benchmarked multiple reinforcement learning algorithms and identified open research challenges.
- The benchmark aims to advance AI research in strategic decision-making under uncertainty.
A team of researchers has introduced FootsiesGym, an open-source reinforcement learning environment tailored for two-player, zero-sum, imperfect-information games. The benchmark is built on HiFight's minimalist 2D fighting game Footsies, which isolates the cyclic and non-transitive strategic interactions typical of fighting game neutral play. Unlike traditional games, FootsiesGym focuses on scenarios where players lack complete information, making it a challenging testbed for AI agents.
The environment includes a vectorized simulator that allows for high-throughput training on standard hardware, ensuring accessibility and reproducibility. This design choice enables researchers to efficiently benchmark reinforcement learning algorithms without requiring specialized infrastructure. The paper accompanying the release details the environment's architecture, evaluates several state-of-the-art RL algorithms, and highlights open research questions in imperfect-information game theory.
Source: FootsiesGym: A Fighting Game Benchmark for Two-Player Zero-Sum Imperfect-Information Games. Read the full piece at the source.
Provides a new, accessible environment for testing RL algorithms in imperfect-information scenarios.
Offers a practical tool for learning about reinforcement learning and game theory.
Expands the toolkit for AI research in strategic decision-making.
- Imperfect-information game
- A game where players do not have full knowledge of the game state or opponent's moves.
- Non-transitive strategies
- Strategies where no single approach dominates all others, creating cyclic advantages.
- Vectorized simulator
- A simulation environment that processes multiple game states simultaneously for efficient training.
AI ResearchAnt Group’s Robbyant Open-Sources LingBot-Vision: A 1B Boundary-Centric Vision Foundation Model for Dense Spatial Perception
![[AINews] Lilian Weng summarizes 35 papers on Harness Engineering for RSI](https://images.weserv.nl/?url=substackcdn.com%2Fimage%2Ffetch%2F%24s_%21L_Ci%21%2Cf_auto%2Cq_auto%3Agood%2Cfl_progressive%3Asteep%2Fhttps%253A%252F%252Fsubstack-post-media.s3.amazonaws.com%252Fpublic%252Fimages%252F603a46c6-cedc-4b38-a660-2fa1d4b3f4ba_1626x1146.png&w=520&fit=cover&q=70&output=webp&dpr=2&we=1&il=1)
AI Research[AINews] Lilian Weng summarizes 35 papers on Harness Engineering for RSI
Model Panic: How Fear of Open-Source AI Is Ceding Ground to China - R Street Institute
Introducing Muse Image and Muse Video - AI at Meta
ELSA3D: Elastic Semantic Anchoring for Unified 3D Understanding and Generation
Scoop: Trump administration lifts restrictions on OpenAI's GPT 5.6 - Axios
The Trump administration has lifted restrictions on OpenAI's GPT 5.6, allowing the company to deploy the model without prior government approval.
Resilience by Design: Preparing Your AI Stack for an Era of Uncertainty - Bain & Company
Bain & Company discusses preparing AI stacks for uncertainty, focusing on resilience by design. This approach aims to help organizations navigate potential AI disruptions.
LLMNVIDIA Releases Audex (Nemotron-Labs-Audex-30B-A3B): A Unified Audio-Text LLM That Preserves the Text Intelligence of Its Backbone
NVIDIA introduces Audex 30B-A3B, a mixture-of-experts model combining audio understanding, speech recognition, translation, TTS, and audio generation while maintaining high text intelligence from its backbone.
Meta Built An AI Detection Tool To ID Images And Video Created With Its New Models - Engadget
Meta has introduced an AI detection tool designed to identify images and videos generated by its latest AI models, aiming to combat misinformation and enhance transparency.
Kalshi traders see slim odds U.S. government will take a stake in OpenAI this year - CNBC
Kalshi prediction markets suggest less than a 5% chance the U.S. government will take a stake in OpenAI before 2025.
Forsyth County Sheriff's Office tests out new humanoid robot, artificial intelligence - wfmynews2.com
Forsyth County Sheriff's Office is piloting a humanoid robot equipped with AI for law enforcement tasks.