Michael Widrich

PhD candidate, Institute for Machine Learning, Johannes Kepler University Linz

[Michael Widrich @MichaelWidrich]

Tackling an extremely massive multiple instance learning problem in immunology with deep learning and attention (modern Hopfield networks!)
Check out our spotlight and poster at #neurips2020! 🙂
Dec 09, 07:10-07:20 (Spotlight), 09:00-11:00 (Poster) AM PST
https://t.co/PwQrdCgqkG

[Marius-Constantin Dinu @DinuMariusC]

Tweaking #OfflineRL methods seems tricky... but do we have enough understanding about the influence of the datasets in use on OfflineRL methods, before speaking of tuning them?
We shed light on the dataset characteristics and their effects on OfflineRL: https://t.co/9vofzydnD9

[Johannes Brandstetter @jo_brandstetter]

Wow, wanna see how to beat CLIP with the new CLOOB? Fantastic work lead by my colleagues @fuerst_andreas and @LizRumetshofer (Sepp Hochreiter's group) applying modern Hopfield networks to image-text data.

Paper: https://t.co/uATlB6nr9D
Blogpost: https://t.co/RLKgNYtL8M

[Philippe A. Robert @PRobertImmodels]

Happy to share our preprint: "A billion synthetic 3D-antibody-antigen complexes enable unconstrained machine-learning formalized investigation of antibody specificity prediction"
https://t.co/56rffLJfwk
Check out Absolut! database+software to simulate more
https://t.co/Y0LZcIE9VW

[PyTorch @PyTorch]

PyTorch 1.9 is here! Highlights include improvements for:
- torch.linalg, torch.special, and Complex Autograd
- Mobile Interpreter
- TorchElastic
- The PyTorch RPC framework
- APIs for model inference deployment
- PyTorch Profiler

See full details👇
https://t.co/pATnZRFQkx

[Max Welling @wellingmax]

Fantastic work by the AMLAB team Johannes Brandstetter @jbrandi6, Rob Hesselink @robdhess, Elise van der Pol @ElisevanderPol and Erik Bekkers @erikjbekkers. We trained a fully steerable equivariant GNN to reach the top of the leaderboard. Congrats team.

[GreiffLab 💻🔬💊 @victorgreiff]

My @airr_community/@TheAbSociety webinar on immune receptor (BCR, TCR) data analysis is now available on-demand for free (including the slides and written answers to all webinar questions; also those we did not have time to answer during the webinar): https://t.co/fymtso04EA

[ELLIS Unit Linz & LIT AI Lab @LITAILab]

In our new paper, we identify and counter domain shifts in machine learning models for COVID-19 diagnosis from blood tests: https://t.co/P0hu4YdiT7

[Geir Kjetil Sandve @SandveGeir]

Very happy to finally release immuneML – an open-source python platform that streamlines development, application, and benchmarking of immune receptor machine learning for predicting immune status or antigen specificity (https://t.co/BmHuJddoel). https://t.co/ALXPCY9tL0 1/10

[Aging Science News @AgingBiology]

Hallmarks and detection techniques of cellular senescence and cellular ageing in immune cells
https://t.co/7SeaNbowkJ

[Johannes Brandstetter @jo_brandstetter]

Our paper "Hopfield Networks is All You Need" is accepted at #ICLR2021. Time to give some talks :) I am very honored to present our research today at the great platform of @ml_collective @savvyRL (https://t.co/vb2n5cMjcL).

[Philippe A. Robert @PRobertImmodels]

Our review is preprinted🥳!

Modelling the Dynamics of T-Cell Development in the Thymus -> we show how computational models have extracted the secret dynamics of thymocytes

Super nice collaboration with @KruegerLab @victorgreiff and Heike Kunze-Schumacher. Happy reading!

[Michael Widrich @MichaelWidrich]

Spotlights Track 15: COVID/Applications/Composition is live now, Poster session is afterwards (18:00 CET) ! Looking forward to seeing you there! :)

[Michael Widrich @MichaelWidrich]

Tackling an extremely massive multiple instance learning problem in immunology with deep learning and attention (modern Hopfield networks!)
Check out our spotlight and poster at #neurips2020! 🙂
Dec 09, 07:10-07:20 (Spotlight), 09:00-11:00 (Poster) AM PST
https://t.co/PwQrdCgqkG

[Michael Widrich @MichaelWidrich]

@tommylees112 glad you like it! :)
I think that figure was actually created by günter in libreoffice, the other ones in the paper are in inkscape 🤓

[Michael Widrich @MichaelWidrich]

Tackling an extremely massive multiple instance learning problem in immunology with deep learning and attention (modern Hopfield networks!)
Check out our spotlight and poster at #neurips2020! 🙂
Dec 09, 07:10-07:20 (Spotlight), 09:00-11:00 (Poster) AM PST
https://t.co/PwQrdCgqkG

[Michael Widrich @MichaelWidrich]

Want to tackle massive MIL problems with DeepRC?
DeepRC github repo: https://t.co/e8iKQwMIGN
More on modern Hopfield networks: https://t.co/RuTuNQYXSz
Our papers: https://t.co/6933K3fZrE, https://t.co/THLRqjvYHz

@bschaefl @jbrandi6 @HRamses2 @LITAILab @milenapavl @victorgreiff

[Michael Widrich @MichaelWidrich]

Does it work?
DeepRC outperformed all compared methods, significantly outperforming the second best method.
DeepRC allowed to retrieve the motifs we implanted in the simulated datasets. For the real dataset, the found sequences correspond to those identified by other publications

[GreiffLab 💻🔬💊 @victorgreiff]

Our review on --- Mining adaptive immune receptor repertoires for biological and clinical information using machine learning --- is online. https://t.co/1JGOQA1YvT. Fun project with @GurYaari and Lindsay Cowell. @airr_community @ireceptor_plus @UiO_LifeSci @RCN_Norway