Skip to content

[Feature]: Evaluate coordinated pretraining for all AutoE2E encoders #156

Description

@riita10069

Problem / motivation

The navigation-input design defers encoder pretraining. Pretraining only the map encoder would leave the rest of the encoder stack under a different initialization strategy.

Proposed solution

Evaluate coordinated pretraining for the camera, semantic-map, route, temporal-history, and world-model encoders. Compare the current initialization with frozen, partially frozen, and end-to-end fine-tuned pretrained variants, and report convergence and downstream KITScenes metrics.

Relevant references: MAE and I-JEPA.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions