Andreas Kipf

@andreaskipf

PhD student in databases at TUM. Interested in ML for systems. Former intern.

Munich, Bavaria
db.in.tum.de/~kipf/
Vrijeme pridruživanja: lipanj 2009.
10 fotografija ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @andreaskipf

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @andreaskipf

  1. Prikvačeni tweet
    1. pro 2019.

    Excited to share our workshop paper on benchmarking learned indexes! Joint work with and Paper: Code:

    1 reply 12 proslijeđenih tweetova 47 korisnika označava da im se sviđa
    Poništi
  2. 1. velj

    Successfully defended my PhD 🥳

    28 replies 4 proslijeđena tweeta 372 korisnika označavaju da im se sviđa
    Poništi
  3. proslijedio/la je Tweet
    27. sij

    New work on multiplexing simple query optimizers at . Learned QO trained in < 1 hour. Come see my short talk this afternoon and stick around for the poster session!

    2 proslijeđena tweeta 24 korisnika označavaju da im se sviđa
    Poništi
  4. proslijedio/la je Tweet
    26. sij

    This is Thomas Neumann: People outside of academia may not know but he is probably the #1 DB systems researcher in the world. He sold his last DBMS (HyPer) to Tableau: He just announced his new DBMS Umbra:

    Thomas Neumann's reactions after Andy tells him about trying to bribe the CMU CS dept chair with fake money.
    86 proslijeđenih tweetova 394 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  5. proslijedio/la je Tweet
    12. pro 2019.

    See/hear researchers present "SOSD: A Benchmark for Learned Indexes" MLSys Workshop 12/13 et al

    1 proslijeđeni tweet 2 korisnika označavaju da im se sviđa
    Poništi
  6. proslijedio/la je Tweet
    5. pro 2019.

    Our paper on multi-dimensional learned index structures got accepted at . Our learned index is up to 3 orders-of-magnitude faster than other state-of-the art multi-dim indexes or sort orders. If you want to learn more the arxiv paper is now online

    20 proslijeđenih tweetova 99 korisnika označava da im se sviđa
    Poništi
  7. proslijedio/la je Tweet
    3. pro 2019.

    Very excited to see this benchmark out there. It also includes the open-source implementation of two learned indexes, RMI (trained top-down) and RS (trained bottom-up), which show very strong overall performance.

    Excited to share our workshop paper on benchmarking learned indexes! Joint work with and Paper: Code:
    2 proslijeđena tweeta 21 korisnik označava da mu se sviđa
    Poništi
  8. proslijedio/la je Tweet
    2. pro 2019.

    SOSD, our new benchmark for learned and traditional index structures, contains 3 real-world datasets & the first publicly available RMI implementation! Check it out: Code:   Paper: Joint work with

    6 proslijeđenih tweetova 21 korisnik označava da mu se sviđa
    Poništi
  9. 9. ruj 2019.

    Check out my talk on cardinality estimation with deep learning

    1 proslijeđeni tweet 11 korisnika označava da im se sviđa
    Poništi
  10. proslijedio/la je Tweet
    29. kol 2019.

    Here's my non-exhaustive checklist I think every good ML-for-DB paper should have, especially query optimization:

    14 proslijeđenih tweetova 47 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  11. proslijedio/la je Tweet
    26. kol 2019.

    is talking about estimating filtered groupby queries using deep learning!

    1 proslijeđeni tweet 5 korisnika označava da im se sviđa
    Poništi
  12. proslijedio/la je Tweet
    1. kol 2019.

    AIDB 2019 workshop is approaching!!! (), (), and Feifei Li () will share their experiences in building next-gen AI-powered DBMS!!! Come and join us in LA!

    5 proslijeđenih tweetova 18 korisnika označava da im se sviđa
    Poništi
  13. proslijedio/la je Tweet
    24. srp 2019.

    Nice practical corrected-for-numerical-overflow trick for arbitrary sized cuckoo filters:

    1 proslijeđeni tweet 2 korisnika označavaju da im se sviđa
    Poništi
  14. proslijedio/la je Tweet
    5. srp 2019.

    DuckDB: A New, Embeddable SQL OLAP DBMS -

    5 proslijeđenih tweetova 11 korisnika označava da im se sviđa
    Poništi
  15. 29. lip 2019.

    Just arrived in Amsterdam for . We‘ll show a demo on cardinality estimation with deep sketches. Would be great to see you around!

    Our demo "Estimating Cardinalities with Deep Sketches" aka learned cardinalities in action has been accepted to with
    Prikaži ovu nit
    15 korisnika označava da im se sviđa
    Poništi
  16. proslijedio/la je Tweet
    23. lip 2019.

    How can we efficiently remove user data from trained ML models to enact the right to be forgotten? We present some initial ideas for simple models in an upcoming abstract for the AIDB workshop at VLDB.

    1 reply 11 proslijeđenih tweetova 46 korisnika označava da im se sviđa
    Poništi
  17. 17. lip 2019.

    New preprint on "DeepSPACE: Approximate Geospatial Query Processing with Deep Learning". We are using an autoregressive model to predict the results of COUNT, AVG, and SUM aggregation queries on the NYC taxi dataset:

    2 proslijeđena tweeta 15 korisnika označava da im se sviđa
    Poništi
  18. 7. lip 2019.

    Our paper "Estimating Filtered Group-By Queries is Hard: Deep Learning to the Rescue" has been accepted to AIDB

    1 proslijeđeni tweet 12 korisnika označava da im se sviđa
    Poništi
  19. proslijedio/la je Tweet
    16. svi 2019.

    New blog post by Thomas Neumann on using spline interpolation as an alternative to learned indexes:

    9 proslijeđenih tweetova 26 korisnika označava da im se sviđa
    Poništi
  20. 25. ožu 2019.

    We've just published the source code of Learned Cardinalities:

    3 proslijeđena tweeta 15 korisnika označava da im se sviđa
    Poništi

@andreaskipf još nije objavio nijedan Tweet.

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·