LAION-AI / LAION-AI/Open-Assistant

Why use L2 regularization in reward model training?

Open
#3,675 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
37.4k
Forks
3.3k
PR merge metrics
No merged PRs in 30d

Description

Hello, respected developers of Open Assistant. @andreaskoepf While studying your reward model training code, I noticed that besides the ranking loss, there is an additional L2 regularization term. What is the purpose of this regularization term? Are there any papers that mention it?
https://github.com/LAION-AI/Open-Assistant/blob/7e40ee313bd327ca069e1d8b38efc371b66dea6f/model/model_training/utils/losses.py#L76

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with model/model_training/utils/losses.py at the linked line and inspect how the reward model ranking loss and L2 term are used. Research the purpose of this regularization in reward model training and identify relevant papers. Done means documenting a clear explanation with supporting citations.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Documentation
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.