Sebastian Blank

[Anton Wiehe @AntonWiehe]

I'm excited to share our newest publication: "Language over Labels": https://t.co/hoHIEFtaCx
CLIP does not transfer well to radiology images. We use contrastive language supervision with a linear probe (CLS+LP) to outperform regular fine-tuning (FT) at varying data sizes.

[inovex GmbH @inovexgmbh]

Information Extraction From Semi-Structuted Data Using Machine Learning, #blog by Pol Zeimet: https://t.co/LnYSxrJUXW #MachineLearning #InformationExtraction

[inovex GmbH @inovexgmbh]

Bei unserem Meetup zum Thema One Shot Learning blieben einige Fragen unbeantwortet. Mai und @_SebastianBlank haben diese beantwortet: https://t.co/fGmS1KtUbC #OneShotLearning #DeepLearning

[Sebastian Blank @_SebastianBlank]

We @inovexgmbh worked with five motivated students to bring an image captioning service from whiteboard to a scalable data product.
https://t.co/eXyucSRT0Q
#computervision #nlp #deeplearning

[inovex GmbH @inovexgmbh]

Kollegin ⁦@MarsaMo1⁩ spricht heute in der FAZ über KI! 🎉🙌 https://t.co/Xsy5PvWvJ8

[Uwe L. Korn @xhochy]

Want to export your conda environment/requirements but don't list all dependencies but only those you directly depend on / have installed explicitly, use `--from-history`

[hardmaru @hardmaru]

Uncertainty Quantification in Deep Learning

A nice overview blog post by @simonbachstein

https://t.co/ZZExu1Vlau

[Pascal Fecht @p_fecht]

For everyone interested in German text summarization systems, have a look at the proceedings of Swisstext 2019 (https://t.co/40v4WzmQKm) to see our (@SbstnBlank, @data_hpz) contributed system description. We used a deep neural net with multilingual BERT embeddings. #SwissText2019

[@[email protected] @berlinbuzzwords]

Would you like to listen to @data_hpz and @SbstnBlank's talk "Querying Elasticsearch with Deep Learning to Answer Natural Language Questions" again? Head over to YouTube right now and just do it (don't let your dreams be dreams): https://t.co/VntBc5mb1J

[inovex GmbH @inovexgmbh]

Heute steigt unser erstes #Meetup im @Karlsruher Technologiepark: Im TechLunch stellt @SbstnBlank seine Masterthesis zu NLP und Elasticsearch vor. 🤖 🔎

[Sebastian Blank @_SebastianBlank]

Check out the slides and video of our talk at #bbuzz this week.
@data_hpz @berlinbuzzwords

https://t.co/oVVzTJusdW
https://t.co/CMadE3Gt4K

[Pascal Fecht @p_fecht]

Great to see the active research in German text summarization at @SwissTextConf. Tomorrow I‘m presenting our (@SbstnBlank @data_hpz) solution with multilingual BERT embeddings for the shared task. @inovexgmbh #SwissText2019

[Lucian Precup @lucianprecup]

Looks like voice assistants and natural language queries are well represented this year @berlinbuzzwords #bbuzz. Deep Learning helps indeed @SbstnBlank @data_hpz .

[inovex GmbH @inovexgmbh]

Neues Format! Wir möchten mit euch die Mittagspause verbringen – inklusive Tech Talk zu einem unserer Fokusthemen sowie Pizza & Networking.
Den Anfang macht @SbstnBlank, der über "Querying Elasticsearch by Asking Questions about Movies" spricht. https://t.co/g8vY26ANaq #Karlsruhe

[inovex GmbH @inovexgmbh]

Price Prediction in Online Car Marketplaces using Natural Language Processing, #blog by @Maxi_Data https://t.co/ruC4jjNrqb #NLProc #MachineLearning #DataScience

[Sebastian Ruder @seb_ruder]

This is a super cool resource: Papers With Code now includes 950+ ML tasks, 500+ evaluation tables (including SOTA results) and 8500+ papers with code. Probably the largest collection of NLP tasks I've seen including 140+ tasks and 100 datasets.
https://t.co/lTAGE7LGZY

[Sebastian Blank @_SebastianBlank]

If you‘re interested, check out our paper: https://t.co/O5ZvhC95BA or our blogpost: https://t.co/8lvCVEOphL about Querying NoSQL with Deep Learning

[Sebastian Blank @_SebastianBlank]

It was a great experience to present our paper at #AAAI2019. Looking forward to see the keynote of @goodfellow_ian today.