> For the complete documentation index, see [llms.txt](https://www.mlcompendium.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://www.mlcompendium.com/types-of-machine-learning/unlearning.md).

# Unlearning

Machine unlearning (MU) refers to the challenge of erasing a data point's influence on the input-output mapping of an ML model.

## Papers

1. (really good) A [survey](https://arxiv.org/abs/2209.02299) of [MU](https://arxiv.org/pdf/2209.02299.pdf).
2. [Existing literature on MU](https://github.com/jjbrophy47/machine_unlearning)
3. [Machine Unlearning: The Right to be Forgotten](https://www.kaggle.com/code/tamlhp/machine-unlearning-the-right-to-be-forgotten#sec:algorithms)
4. (really good) [Awesome MU on Github](https://github.com/tamlhp/awesome-machine-unlearning?tab=readme-ov-file#type-image) ([website](https://awesome-machine-unlearning.github.io/))- a collection of academic articles, published methodology, and datasets on the subject of machine unlearning. model agnostic, intrinsic, and data-driven approaches, evaluation metrics, and datasets.

<figure><img src="/files/LlXdAqY3rPnbjdnwcUR1" alt=""><figcaption></figcaption></figure>

5. Who's Harry Potter? Approximate Unlearning in LLMs. [Arxiv](https://arxiv.org/abs/2310.02238), [paper](https://browse.arxiv.org/pdf/2310.02238), [Microsoft](https://www.microsoft.com/en-us/research/project/physics-of-agi/articles/whos-harry-potter-making-llms-forget-2/), [medium](https://pub.towardsai.net/who-is-harry-potter-inside-microsoft-researchs-fine-tuning-method-for-unlearning-concepts-in-llms-33dfe8e742a9)&#x20;
6. [fast yet effective MU](https://arxiv.org/pdf/2111.08947.pdf)
7. [one shot MU](https://arxiv.org/pdf/2201.05629.pdf)
8. [A review on MU](https://link.springer.com/article/10.1007/s42979-023-01767-4), Zhang et al.
9. [Machine Un-learning: An Overview of Techniques, Applications, and Future Directions](https://link.springer.com/article/10.1007/s12559-023-10219-3), Siva et al.
10. [Machine Unlearning](https://arxiv.org/abs/1912.03817)

## Medium

1. [A fresh perspective on machine unlearning, with a real-world solution!](https://medium.com/@aliborji/a-fresh-perspective-on-machine-unlearning-with-a-real-world-solution-203821dd01c0) a solution that uses three of the following approaches. Data Augmentation, Weight Decay, Fine-Tuning, Selective Retraining, and Neural Architecture Modifications.
2. What is MU? Part [1](https://medium.com/@choquette.christopher/what-is-machine-unlearning-pt-1-933ff53dc9a6), [2](https://medium.com/@choquette.christopher/how-to-do-machine-unlearning-pt-2-ae32cb6ca2f1)

## GitHub

1. [search results](https://github.com/search?q=unlearning\&type=repositories)
2. This [repository](https://github.com/cleverhans-lab/machine-unlearning) contains the core code used in the SISA experiments of our [Machine Unlearning](https://arxiv.org/abs/1912.03817) paper along with some example scripts.
3. [Implementations of various data deletion methods.](https://github.com/ChrisWaites/data-deletion?tab=readme-ov-file) \
   [Evaluation Doc.](https://docs.google.com/document/d/14B_aLihLTNE7a2yRQHNRRVwvSOkttakVYFhlayZBNkE/edit)
4. This [repository](https://github.com/shash42/Evaluating-Inexact-Unlearning/tree/master) contains the code used in our experiments of our paper on [Evaluating Machine Unlearning](https://arxiv.org/abs/2201.06640) in the src/ folder along with some sample scripts in the scripts/ folder.
5. #### [This](https://github.com/meghdadk/SCRUB) is a Python implementation of "Towards Unbounded Machine Unlearning"
6. #### [data deletion](https://github.com/ChrisWaites/data-deletion?tab=readme-ov-file)

## Community

1. NeurIPS 2023 [Kaggle](https://www.kaggle.com/competitions/neurips-2023-machine-unlearning/leaderboard) - Machine Unlearning Erase the influence of requested samples without hurting accuracy


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://www.mlcompendium.com/types-of-machine-learning/unlearning.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.