MultiAgentLearning/playground

Help wanted

Open

#202 opened on Mar 5, 2019

View on GitHub
 (6 comments) (0 reactions) (0 assignees)Python (728 stars) (215 forks)batch import
enhancementhelp wanted

Description

Here are three implementations that the community would really appreciate:

  1. Add in the flames and the direction of movement to the observation: This is here because the top competitors all recreated it.

  2. Make a history function that returns all the states that this agent experienced and an ability to start from any of these states: This is here because it helps a lot with debugging agents and understanding where they are going wrong.

  3. Make a doc that details how agents move: This is here because a lot of folks ask about it and it's an imperative to understand before building an agent in earnest.

Contributor guide

Tech stack
python
Domain
machine learningai
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
needs investigation
Prerequisites
Pythonmulti agent systems
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
40
Research direction
The issue requests three features: (1) add flames and movement direction to observations, (2) implement a history function to replay agent states, and (3) document agent movement. For feature 1, examine the observation rendering code in the environment module. For feature 2, consider a data structure to store states and provide a replay mechanism. For feature 3, document the movement logic visible in the environment's step function. Check existing comments for any partial implementations or discussions.