BigQuery now supports the contribution analysis model (available in public preview)—allowing you to find key drivers of unanticipated changes. This model can effectively scale by pruning the search space through a support threshold.
Here's how it works ↓ https://t.co/fJRSNayuhz
Whether you're the first in your family to attend college at UW or elsewhere, today we #CelebrateFirstGen, your success and your presence! #UWFirstGen
https://t.co/qcgYpuJzwL
Today I'll be giving a talk about our work "SLAOrchestrator: Reducing the Cost of Performance SLAs for Cloud Data Analytics" at USENIX ATC #atc18. Come see my talk or visit my poster! 🎉
More than 70 refereed papers will be presented at @usenix#atc18 in Boston, July 11-13! Topics include SSDs, The Network, Storage, Transactions, Data Center/Machine Learning, Key/Value Storage, and more. View the full program and register now: https://t.co/xeAJSKOO86
#deem18 great talk from Jennifer Ortiz @selectfromjenny , "Learning State Representations for Query Optimization with Deep Reinforcement Learning." DRL can learn competitive cardinality estimates as rich vector embeddings.
Check out their paper: https://t.co/nii6NMmiEE
Excited to present our early work "Learning State Representations for Query Optimization with Deep Reinforcement Learning" at DEEM this year! @deem_workshop@SIGMOD2018 https://t.co/0Iqwl4ZOlV
Should the reviewers of a piece of scientific work be made aware of the identify of the authors? Check out the results of a controlled experiment that compares reviewer behavior under single-blind and double-blind review. https://t.co/dpZAZ1ldgR