Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Discussion on Future Directions #90

Open
maumueller opened this issue Dec 5, 2021 · 5 comments
Open

Discussion on Future Directions #90

maumueller opened this issue Dec 5, 2021 · 5 comments

Comments

@maumueller
Copy link
Collaborator

Dear all,

<tl;dr> Please add your thoughts on the future of this benchmark!

Thank you very much for participating in our NeurIPS'21 competition. The competition will end with an event on Dec 8, and you can find the timeline for this event on https://big-ann-benchmarks.com/. We hope many of you will be able to participate!

The last part of the event will be an open discussion among the participants for future directions of this competition. As organizers we have already identified some points we would like to discuss and potentially include in a future version of the benchmark.

  1. Filtered ANNS: can you support ANNS queries which allow filters like date range, author or some combination of attributes. This would look like a simple SQL + ANNS query.

  2. Streaming ANNS: Can algorithms be robust to insertions and deletions. Here we have a strong baseline (fresh-diskann: https://arxiv.org/abs/2105.09613)

  3. Out of distribution queries: this is already a problem with T2I and we can imagine various variations

  4. Better vector compression: Most approaches use some variant of product quantization as vector compression, but can we get more accurate estimation, maybe at the price of more expensive decoding?

Please let us know what you think about these topics, and add your own!

Thanks!

@fuchun-wang
Copy link
Contributor

Track1 is a general, classic setting track. It's easy to follow for some fresh learner except the huuuuuuuge dataset, and loooooong running time. Maybe we can build a mini-track1 and benchmark tools to verify ideas.

Really thank you for the great event.

@NJU-yasuo
Copy link
Contributor

I wonder if it's free to use any algorithm in T1/T2 instead of being confined to FAISS/DISKANN ?

@fuchun-wang
Copy link
Contributor

Use any algorithm you want is ok.

@fuchun-wang
Copy link
Contributor

fuchun-wang commented Dec 16, 2021 via email

@markwhen
Copy link

Hi @maumueller , is there future competition for these new topics?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants