IdeaCredIdeaCred

chowfi/JEPA-world-model

42

Train a JEPA world model on a set of pre-collected trajectories from an environment involving an agent in two rooms.

What's novel

Train a JEPA world model on a set of pre-collected trajectories from an environment involving an agent in two rooms.

Code Analysis

11 files read · 4 rounds

Trains a JEPA-style self-supervised world model (CNN encoder + recurrent CNN predictor) on 2D agent-in-maze trajectories to predict future latent states conditioned on actions, then evaluates via a frozen-encoder probing head that regresses agent (x,y) location.

Strengths

The recurrent predictor architecture that rolls spatial latents forward one action at a time is a reasonable design choice, and the project demonstrates a working end-to-end pipeline from training to probing evaluation. The DDP adaptation, while incomplete, shows awareness of multi-GPU training patterns.

Weaknesses

The training loop contains a critical bug where the VICReg regularization terms are computed but never used in the actual loss, the target encoder is updated via EMA but never actually used to produce prediction targets (contradicting the README), and the hardcoded 4×4 spatial dimensions will crash on the actual 65×65 data. There are zero tests, duplicated code (Prober in two files), TODO comments left in production code, and the DDP file is an incomplete copy-paste with dummy data paths.

Score Breakdown

Innovation
5 (25%)
Craft
20 (35%)
Traction
17 (15%)
Scope
56 (25%)

Signal breakdown

Innovation

Not Fork+1
Code Novelty+1
Concept Novelty+2

Craft

Ci-3
Tests-5
Polish+0
Releases-1
Has License+0
Code Quality+8
Readme Quality+4
Recent Activity+7
Structure Quality+5
Commit Consistency+0
Has Dependency Mgmt+5

Traction

Forks+6
Stars+6
Hn Points+0
Watchers+3
Early Traction+0
Devto Reactions+0
Community Contribs+2

Scope

Commits+3
Languages+5
Subsystems+5
Bloat Penalty+0
Completeness+6
Contributors+7
Authored Files+12
Readme Code Match+3
Architecture Depth+3
Implementation Depth+8

Evidence

Commits

5

Contributors

3

Files

53

Active weeks

2

TestsCI/CDREADMELicenseContributing

Repository

Language

Python

Stars

3

Forks

1

License