Listen to AMPLab’s (Algorithms Machines People) Daniel Crankshaw talk about Velox! What’s Velox? Velox is a new component of the Berkeley Data Analytics Stack that addresses the critical missing component of current analytics process: the deployment and serving of models at scale. Filmed March 11, 2015 at Marin Software SF, CA. Thanks to everyone who came out to this event! Here's a link to Dan Crankshaw's slides: http://www.slideshare.net/dscrankshaw/velox-at-sf-data-mining-meetup Here are a few of the papers Dan mentions in the presentation: "A Contextual-Bandit Approach to Personalized News Article Recommendation," Lihong Li et al (http://www.research.rutgers.edu/~lihong/pub/Li10Contextual.pdf) "LASER: a scalable response prediction platform for online advertising" Deepak Agrawal et al. from LinkedIn (http://dl.acm.org/citation.cfm?id=2556252) And the two on fast top-k: "Fast top-k similarity queries via matrix compression" Yucheng Low et al. (http://research.microsoft.com/pubs/171030/topk.pdf) "Asymmetric LSH (ALSH) for Sublinear Time Maximum Inner Product Search (MIPS)" Shrivastava et al. (http://papers.nips.cc/paper/5329-asymmetric-lsh-alsh-for-sublinear-time-maximum-inner-product-search-mips.pdf)

 917      4