instadeepai/Mava

[BUG]Jumanji RWARE reward shape mismatch

Open

#1,110 opened on Oct 24, 2024

View on GitHub
 (1 comment) (0 reactions) (0 assignees)Python (122 forks)auto 404
buggood first issue

Repository metrics

Stars
 (922 stars)
PR merge metrics
 (PR metrics pending)

Description

We found that in the 4ag RWARE task, env.reward_spec().generate_value() gives something of shape (), but the reward that environment outputs is actually of shape (num_agents).

This is curious because I believe the scenario should output a reward of shape ().

Maybe it's a wrapper thing?

Contributor guide

Research direction
Investigate the reward function in the 4ag RWARE task and trace through the environment's step function to identify why reward.spec() shape differs from actual reward shape. Check if a wrapper modifies the reward.
Tech stack
python
Domain
machine learningai
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
PythonJAXreinforcement learning
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
60