Issues

• [EVAL] Ablation study runner + charts

  sebcheung/Competitive-MARL-Racing#17 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 5 - Evaluation and Grand Prix

• [EVAL] Tournament runner

  sebcheung/Competitive-MARL-Racing#16 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 5 - Evaluation and Grand Prix

• [EVAL] Elo rating system

  sebcheung/Competitive-MARL-Racing#15 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 5 - Evaluation and Grand Prix

• [REWARD] Agent archetype behavioral priors

  sebcheung/Competitive-MARL-Racing#14 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 4 - Reward Shaping and Agent Archetypes

• [REWARD] Modular reward function with ablation toggles

  sebcheung/Competitive-MARL-Racing#13 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 4 - Reward Shaping and Agent Archetypes

• [MARL] League opponent sampler

  sebcheung/Competitive-MARL-Racing#12 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 3 - Multi-Agent and Self-Play

• [MARL] Self-play loop with agent snapshot system

  sebcheung/Competitive-MARL-Racing#11 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 3 - Multi-Agent and Self-Play

• [MARL] Multi-agent environment extension

  sebcheung/Competitive-MARL-Racing#10 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 3 - Multi-Agent and Self-Play

• [PPO] Single-agent training loop + TensorBoard logging

  sebcheung/Competitive-MARL-Racing#9 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 2 - PPO Implementation

• [PPO] Implement clipped surrogate objective

  sebcheung/Competitive-MARL-Racing#8 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 2 - PPO Implementation

• [PPO] Implement GAE and rollout buffer

  sebcheung/Competitive-MARL-Racing#7 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 2 - PPO Implementation

• [PPO] Implement Actor-Critic network

  sebcheung/Competitive-MARL-Racing#6 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 2 - PPO Implementation

• [ENV] Manual keyboard control test

  sebcheung/Competitive-MARL-Racing#5 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 1 - Environment Foundation

• [ENV] Define observation and action spaces

  sebcheung/Competitive-MARL-Racing#4 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 1 - Environment Foundation

• [ENV] Build circuit track with boundary collision

  sebcheung/Competitive-MARL-Racing#3 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 1 - Environment Foundation

• [ENV] Implement Box2D vehicle physics

  sebcheung/Competitive-MARL-Racing#2 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 1 - Environment Foundation

• [ENV] Set up Gymnasium custom environment scaffold

  sebcheung/Competitive-MARL-Racing#1 · Filter by author sebcheung opened on Feb 26, 2026 · Milestone 1 - Environment Foundation