DC@KTH (@dcatkth) profile picture

Distributed Computing Group at KTH

Joined June 2019
Photo by DC@KTH (@dcatkth)Photo by DC@KTH (@dcatkth)Photo by DC@KTH (@dcatkth)Photo by DC@KTH (@dcatkth)Photo by DC@KTH (@dcatkth)Photo by DC@KTH (@dcatkth)
DC@KTH retweeted
🔥 Lowkey Goated When #FewShotNER Is The Vibe 🤩 Check out this awesome paper by Amirhossein Layegh et al. for Contrastive-based Prompt Tuning: deepai.org/publication/contr… @APayberah
2
8
1,580
@DataCloud2020 Informal Review - 15 Nov 2021 Read more here ⬇ datacloudproject.eu/whats-ne…
1
1
DataCloud Technical Meeting in Stockholm 17-19 November 2021 Read more here ⬇ datacloudproject.eu/whats-ne…
6
6
Nice to see deeper collaboration between @logicalclocks and @NVIDIAAI
On Sept 16, join @logicalclocks and NVIDIA to learn how to identify and prevent #financialfraud and money laundering using #GPU-accelerated deep #anomalydetection methods. Register today: nvda.ws/3zks3Qj
2
8
Feature Store Meetup w/ Airbnb, Logical Clocks & Kaskada Tuesday, December 15, 2020 at 12:00 PM Absolutely Free Online us02web.zoom.us/j/8478660804…
2
1
The inaugural Global Feature Store meetup is taking place on the 15th Dec at 9am PT with @Airbnb , @KaskadaInc , and @logicalclocks - meetup.com/Feature-Stores-fo…
EuroMLSys: the 1st workshop on Machine Learning and Systems, co-located with EuroSys'21. A good venue for you who work on the intersection of Machine Learning and Computer Systems. euromlsys.eu/ @euromlsys @EuroSys_conf
1
3
13
Tomorrow, myself and @sinash93 will be talking at the Data+AI summit about how we can parallelize ablation studies on Spark (including using GPUs, if needed) using the #maggy framework on #hopsworks. databricks.com/session_eu20/…
5
29
Want to see the new version of the Hopsworks Feature Store in action? Then join our live demo at the Data+AI Summit tomorrow and Thursday.
This Post is from an account that no longer exists. Learn more
2
2
The feature store for ML has been the must-have piece of ML infrastructure for 2020. Its successor as shiny-new-toy-for-ML will be end-to-end provenance in pipelines. We are the first to do "implicit provenance" in infrastructure - no need to re-write your code. @alex_ormenisan
This Post is from an account that no longer exists. Learn more
1
3
We are looking for a Postdoc to work with us @dcatkth, on Decentralized DL/RL, Generalization in RL, and AutoML. For more information (and to apply!) please visit dcatkth.github.io/vacancies/ . Application deadline: August 17th.
Find out what "implicit provenance" for ML Pipelines is, how we added support in @hopsworks, and how it helps with debugging and governing ML pipelines. By @alex_ormenisan . @usenix #OpML20 usenix.org/conference/opml20…
5
12
Find out what an oblivious training function is and how it enables you to write one training function that runs both on your laptop on Jupyter or on a 100 GPU cluster by @morimeister and myself and the @SparkAISummit from last week : youtube.com/watch?time_conti…
I'm looking for a postdoc in our group at @dcatkth . Our research is going to be on: - Decentralized DL/RL - Generalization in RL - Automated ML If you are interested, please apply here: tinyurl.com/y6uspqhs
27
39
I am often on calls with teams who are deciding between building and buying a feature store. So many calls, that I wrote a blog on how to build your own #featurestore. @logicalclocks @hopsworks logicalclocks.com/blog/how-t…
1
3
16
featurestore.org/ is an amazing overview of Feature Stores with videos, links and summaries from @dcatkth
1
3
Updated feature store reference doc with tons of external resources and links by @dcatkth featurestore.org/
1
1
buff.ly/2yFdxIR This is a great collection of talks about the variety of ways that organizations build and manage their feature stores. @dcatkth
1
3