Release 1.5.0 · proroklab/VectorizedMultiAgentSimulator Football revamp We have just revamped VMAS football! New rewards, observations, options for physical differences and formation spawning and value function plotting. Try it out for some intense N v...

💡 If you're working on AI for collective decision-making, now is the time to try it out!
🔗 Full release: github.com/proroklab/VectorizedMultiAgentSimulator/releases/tag/1.5.0
👉 Try it out & let us know what you think! 🚀

6️⃣ Self-Play Ready
🔄 Fully mirrored obs & rewards for two-team setups

5️⃣ Kicking Mechanics
⚽ Enable kicking & rotation actions for advanced control

4️⃣ Formation Spawning
📌 Spawn agents randomly or in structured formations

3️⃣ Optional Physical Differences
⚡ Try new setups with role-based agent

2️⃣ Tunable Heuristic Opponents
🎯 Adjust speed, decision-making & precision (0-1)
⚖️ Fair benchmarking for RL algorithms

1️⃣ New Rewards System
✅ Sparse rewards for scoring
✅ Dense rewards for ball movement & exploration
✅ Fully configurable for curriculum learning

🚀 VMAS Football just got a major upgrade! ⚽🔥

We've revamped it to make continuous control with multi-agent RL even more dynamic & competitive.

Link: github.com/proroklab/VectorizedMultiAgentSimulator/releases/tag/1.5.0

Here’s what’s new: 🧵👇

#AI #ReinforcementLearning #MARL #MachineLearning