Zobrazeno 1 - 10
of 91
pro vyhledávání: '"LIBERTY, EDO"'
Maximum inner product search (MIPS) over dense and sparse vectors have progressed independently in a bifurcated literature for decades; the latter is better known as top-$k$ retrieval in Information Retrieval. This duality exists because sparse and d
Externí odkaz:
http://arxiv.org/abs/2309.09013
Maximum Inner Product Search or top-k retrieval on sparse vectors is well-understood in information retrieval, with a number of mature algorithms that solve it exactly. However, all existing algorithms are tailored to text and frequency-based similar
Externí odkaz:
http://arxiv.org/abs/2301.10622
Autor:
Liberty, Edo
This paper provides a one-line proof of Frequent Directions (FD) for sketching streams of matrices. The simpler proof arises from sketching the covariance of the stream of matrices rather than the stream itself.
Externí odkaz:
http://arxiv.org/abs/2202.01780
Autor:
Krishnan, Aditya, Liberty, Edo
This paper suggests the use of projective clustering based product quantization for improving nearest neighbor and max-inner-product vector search (MIPS) algorithms. We provide anisotropic and quantized variants of projective clustering which outperf
Externí odkaz:
http://arxiv.org/abs/2112.02179
Estimating ranks, quantiles, and distributions over streaming data is a central task in data analysis and monitoring. Given a stream of $n$ items from a data universe equipped with a total order, the task is to compute a sketch (data structure) of si
Externí odkaz:
http://arxiv.org/abs/2004.01668
Approximating quantiles and distributions over streaming data has been studied for roughly two decades now. Recently, Karnin, Lang, and Liberty proposed the first asymptotically optimal algorithm for doing so. This manuscript complements their theore
Externí odkaz:
http://arxiv.org/abs/1907.00236
Random projections (RP) are a popular tool for reducing dimensionality while preserving local geometry. In many applications the data set to be projected is given to us in advance, yet the current RP techniques do not make use of information about th
Externí odkaz:
http://arxiv.org/abs/1906.09489
Autor:
Karnin, Zohar, Liberty, Edo
This paper defines the notion of class discrepancy for families of functions. It shows that low discrepancy classes admit small offline and streaming coresets. We provide general techniques for bounding the class discrepancy of machine learning probl
Externí odkaz:
http://arxiv.org/abs/1906.04845
To make deep neural networks feasible in resource-constrained environments (such as mobile devices), it is beneficial to quantize models by using low-precision weights. One common technique for quantizing neural networks is the straight-through gradi
Externí odkaz:
http://arxiv.org/abs/1810.00861
Publikováno v:
Journal of the ACM; Oct2023, Vol. 70 Issue 5, p1-48, 48p