Faster Maximum Inner Product Search in High Dimensions · HackerPapers discussions

Faster Maximum Inner Product Search in High Dimensions — mentioned by motiwari

Fri, 07 Apr 2023 23:16:30 GMT

Right! Though to clarify a few nits: we use successive elimination instead of successive halving. And we talk about the Maximum Inner Product Search problem (very similar to NN problem) in our followup work: https://ar5iv.org/abs/2212.07551

Paper and discussion history. A mention is not an endorsement.

Faster Maximum Inner Product Search in High Dimensions — mentioned by eternalban

Wed, 05 Apr 2023 20:32:01 GMT

Thanks for sharing your great work! [Try this tool from arxiv labs (just replace the x in original url with a 5)] https://ar5iv.labs.arxiv.org/html/2006.06856 https://ar5iv.labs.arxiv.org/html/2212.07551

Paper and discussion history. A mention is not an endorsement.

Faster Maximum Inner Product Search in High Dimensions — mentioned by motiwari

Wed, 05 Apr 2023 19:30:18 GMT

Really funny that you mention that! Some of our more recent work focuses on using adaptive sampling techniques in approximate-nearest-neighbor search (actually, the related problem of maximum inner product search: https://ar5iv.org/abs/2212.07551). We definitely think that our approach could be used to make an index structure for ANN search directly, for example in conjunction with Hierarchical Navigable Small World approaches.

Paper and discussion history. A mention is not an endorsement.