Fredrik K. Gustafsson

Postdoc at IBME in Oxford. Machine learning for healthcare. I'm more active on

[Fredrik K. Gustafsson @fregu856]

My year of reading in 2024: https://t.co/mYR8v8lsxE

I read 99 papers in 2024. Complete list: https://t.co/PLqCgGKK3o

Top 15 favorite papers that I found particularly interesting and/or well-written (in alphabetical order):

[Fredrik K. Gustafsson @fregu856]

New preprint: Evaluating Computational Pathology Foundation Models for Prostate Cancer Grading under Distribution Shifts

https://t.co/M9Ra3Ldeen

[Fredrik K. Gustafsson @fregu856]

New preprint: Evaluating Deep Regression Models for WSI-Based Gene-Expression Prediction.

https://t.co/n7ilrrwDuH

[Daniel Gedon @danigedon]

Go and work with Antônio! 🤗
He’s an incredible supervisor. Smart, full of ideas and fun to work with. Doing a PhD with him cannot go wrong!
If you consider applying and have questions, hit me up!

[Fredrik K. Gustafsson @fregu856]

Our paper "Evaluating Regression and Probabilistic Methods for ECG-Based Electrolyte Prediction" has been published in Scientific Reports.

Paper: https://t.co/AYxA9kS8d9
Code: https://t.co/ajfNlYAPm2

Work done together with @danigedon, @ahortaribeiro and Uppsala collaborators.

[Fredrik K. Gustafsson @fregu856]

One year ago, I wrote a blog post called "The How and Why of Reading 300 Papers in 5 Years", and now I recently reached 400 read papers on the Github repository I use to track and organize my reading: https://t.co/PLqCgGKcdQ

[Fredrik K. Gustafsson @fregu856]

3. A Foundational Multimodal Vision Language AI Assistant for Human Pathology:
https://t.co/ZAA1e7mXEu

[Fredrik K. Gustafsson @fregu856]

Curious about the latest developments within ML for computational pathology? I can definitely recommend reading these 3 papers from the @AI4Pathology lab:

1. Towards a General-Purpose Foundation Model for Computational Pathology (Nature Medicine, 2024):
https://t.co/zxGbesdJ4h

[Fredrik K. Gustafsson @fregu856]

2. A Visual-Language Foundation Model for Computational Pathology (Nature Medicine, 2024):
https://t.co/F9s2dl0lQA

[Fredrik K. Gustafsson @fregu856]

I recently presented our paper "How Reliable is Your Regression Model's Uncertainty Under Real-World Distribution Shifts?" in the @RISEsweden ML seminars.

Paper: https://t.co/oeU27wj8PX
Code: https://t.co/dettfQJUVY
Slides: https://t.co/vDIdugmHgh
Video: https://t.co/P4jdEJbsOU

[Fredrik K. Gustafsson @fregu856]

Our follow-up paper "Photo-Realistic Image Restoration in the Wild with Controlled Vision-Language Models" will also be presented at a CVPR workshop this summer.

Paper: https://t.co/XWpmyU6F8U
Code: https://t.co/40ugzMWbPU

[Fredrik K. Gustafsson @fregu856]

Our paper "Controlling Vision-Language Models for Multi-Task Image Restoration" will be presented at #ICLR2024 next week.

Paper: https://t.co/0HMsmRKEF4
Code: https://t.co/40ugzMWbPU
Project page: https://t.co/nni7nu4Vrx

Work lead by @ZiweiRo, with collaborators in Uppsala.

[Fredrik K. Gustafsson @fregu856]

Our paper "Controlling Vision-Language Models for Multi-Task Image Restoration" will be presented at #ICLR2024 next week.

Paper: https://t.co/0HMsmRKEF4
Code: https://t.co/40ugzMWbPU
Project page: https://t.co/nni7nu4Vrx

Work lead by @ZiweiRo, with collaborators in Uppsala.

[Fredrik K. Gustafsson @fregu856]

@nikitadurasov I think it could work well as the textbook for a first deep learning course, which follows a course on more traditional ML (e.g. https://t.co/FrTLsXPThi). Then you could have specialized courses on CV, RL, NLP etc.

[Fredrik K. Gustafsson @fregu856]

Just finished reading the "Understanding Deep Learning" book: https://t.co/hDs7tNqiRF

Chapters that I particularly liked / found useful: 1, 2, 8, 9, 12, 14, 15, 17, 18, 21.

It contains a lot of neat/illustrative figures that really help to explain various concepts and methods.

[Fredrik K. Gustafsson @fregu856]

Quite enjoyed reviewing the batch of 6 papers I was assigned for #ICML2024, probably the highest overall quality/relevance for any conference I've done so far.

[Fredrik K. Gustafsson @fregu856]

@ElNahhasOSM @karolinskainst Please send me an email! [email protected]

[Fredrik K. Gustafsson @fregu856]

Two months ago, I defended my thesis "Towards Accurate and Reliable Deep Regression Models".

Pdf: https://t.co/YmPt6hMvZX
Slides: https://t.co/SI4HN1r7fP
Video: https://t.co/04dx6fMPsA

I also joined @karolinskainst as a postdoc, working on ML for computational pathology.

[Fredrik K. Gustafsson @fregu856]

@Uberstedt Tack!

[Fredrik K. Gustafsson @fregu856]

@Abdo_Eldesokey Thank you!