katharina

PhD student @cpjku @jkulinz. 💻🎧🎼🎸🏑

[katharina @katxiip]

Accepted @ismir2021! 🥳 Check out the preprint of our paper now (https://t.co/BJ3ia5FsEs) 👇

[Verena Praher @veroamilbe]

I am happy to announce that our latest work "On the Veracity of Local, Model-agnostic Explanations in Audio Classification: Targeted Investigations with Adversarial Examples" (together w/ @katxiip, Arthur Flexer and Gerhard Widmer from @cpjku) has been accepted to @ismir2021 🎉🥳

[Institute of Computational Perception at JKU Linz @cpjku]

We are proud to announce that our institute won two tasks in the 2024 edition of @DCASE_Challenge
🥇Task 4: Sound Event Detection with Heterogeneous Training Dataset
🥇Task 8: Language-based Audio Retrieval
Check out the DCASE website for more https://t.co/Gu8hmM6Egn
#MIR #DCASE

[Günter Klambauer @gklambauer]

Happy to share the PhD thesis of @ml_hoedt :

"On signal propagation theory in neural networks"

https://t.co/UXeYrWUNtT

[Pieter-Jan Hoedt @ml_hoedt]

This time tomorrow I will be presenting our work on initialising input-convex networks at the #NeurIPS23 poster session!

links: https://t.co/uKNIF8rm8U

Do not hesitate to come by for a chat!

[TISMIR journal @TismirJ]

NEW: Dataset paper by Silvan David Peter, @carloscancino, @fosfrancesco, Andrew Philip McLeod, Florian Henkel, Emmanouil Karystinaios & @GerhardWidmer: "Automatic Note-Level Score-to-Performance Alignments in the ASAP Dataset",largest fully note-aligned ds
https://t.co/36horaeBJ7

[NeurIPS Conference @NeurIPSConf]

We are happy to announce the #NeurIPS2023 Creative AI Track. This track will invite ML researchers to showcase their work during the conference in the form of visual, language, musical, and performing arts. Proposals are due by *June 15*.

https://t.co/N3M1jsBYWx

[katharina @katxiip]

Do we make valid inferences in our MIR experiments?

If you're interested in learning more about validity in MIR, check out the latest work of my colleague Arthur Flexer together with Bob L. T. Sturm, now available on arXiv: https://t.co/sRnl9Tosv4

[katharina @katxiip]

2x @cpjku, going strong again 🤓🥳

[ISMIR Conference @ISMIRConf]

ISMIR Awards! This year's Best Reviewer Awards are for: Katharina Hoedt, @valktalk, Maximilian Schmit, Reinier de VAlk, Silvan Peter, and Thomas Pratzlich. Thank you to all our reviewers and meta-reviewers!

[katharina @katxiip]

@veroamilbe You can also check out the code here: https://t.co/JIMxo1lwpD

[katharina @katxiip]

Can we get one step closer to more human-friendly explanations for music classifiers?

We investigated this in our latest work (w/ Francesco Foscarin, @veroamilbe, Arthur Flexer and Gerhard Widmer), accepted at #ismir2022 🥳

Check out the preprint now: https://t.co/7TpL4TALi2

[katharina @katxiip]

If you're attending SMC-22, check out our poster tomorrow during Paper Session 5 (Friday 10:00 UTC+1) 😎

[katharina @katxiip]

New name, new paper 🙃 Our work on "Defending a Music Recommender Against Hubness-Based Adversarial Attacks" (w/ Arthur Flexer and Gerhard Widmer @cpjku) was recently accepted at SMC-22: https://t.co/x8Feo3ykfO 🥳

[arXiv Sound @ArxivSound]

``Defending a Music Recommender Against Hubness-Based Adversarial Attacks. (arXiv:2205.12032v1 [https://t.co/3pcQCkeyAA]),'' Katharina Hoedt, Arthur Flexer, Gerhard Widmer, https://t.co/k6qWHqftxO

[katharina @katxiip]

New name, new paper 🙃 Our work on "Defending a Music Recommender Against Hubness-Based Adversarial Attacks" (w/ Arthur Flexer and Gerhard Widmer @cpjku) was recently accepted at SMC-22: https://t.co/x8Feo3ykfO 🥳

[katharina @katxiip]

If you ever wondered about (Batch) Normalisation & what happens if you try to get rid of it - or simply want to read a very nice blog-post... 🙃

[Pieter-Jan Hoedt @ml_hoedt]

In case you missed it: I am one of the contributors to the first ICLR blog post track this year! I (tried to) write about Normalizer-Free networks (https://t.co/P6AbpDYlQA) and what they can and can not tell us about normalization:

https://t.co/DsJTo80K4v

[Pieter-Jan Hoedt @ml_hoedt]

I'm officially one of the pioneers of published blog posts now: https://t.co/116rmcFuCe

[Verena Praher @veroamilbe]

Now also available as a nice looking HTML page thanks to https://t.co/NRFH3CxhSI!
https://t.co/PmyUheUV7a
https://t.co/5NmAt25hzM

[TISMIR journal @TismirJ]

NEW: Research paper by ArthurFlexer, Taric Lallai and Katja Rašl: "On Evaluation of Inter- and Intra-Rater Agreement in Music Recommendation" presents 2 user studies in quantification of general similarity between pieces.
https://t.co/qgUy6hGDRZ

[Verena Praher @veroamilbe]

3 people from @cpjku 🎉🎉🎉

[katharina @katxiip]

Check out our poster and join us for a discussion @ismir2021 poster session 1(B) in (less than) an hour (10:30 UTC+1) 👋

[katharina @katxiip]

Accepted @ismir2021! 🥳 Check out the preprint of our paper now (https://t.co/BJ3ia5FsEs) 👇

[Verena Praher @veroamilbe]

I am happy to announce that our latest work "On the Veracity of Local, Model-agnostic Explanations in Audio Classification: Targeted Investigations with Adversarial Examples" (together w/ @katxiip, Arthur Flexer and Gerhard Widmer from @cpjku) has been accepted to @ismir2021 🎉🥳

[Verena Praher @veroamilbe]

Visit our poster in Poster Session 1 (A) in a bit more than hour! (I hope I'll not fall asleep 🥱)

[Pieter-Jan Hoedt @ml_hoedt]

Upon request I recorded a presentation that I was invited to give at the AIDD workshop, past Wednesday.

It’s about tricks of the trade, so I hope there is something in there for someone out there.

And yes, #Vimeo was the easiest way to share the video:
https://t.co/YDTG7KW3E4