this paper's nuts. for sentence classification on out-of-domain datasets, all neural (Transformer or not) approaches lose to good old kNN on representations generated by.... gzip https://t.co/6eZiXlJxOX
I am thrilled to share the news about our latest publication!
The findings from our study suggest that maintaining healthy longevity involves more than just physical activity; it also requires the alignment of our circadian rhythm.
#digitalbiomarkers #…https://t.co/gki5pdGqle
We are thrilled to present our published study protocol on one of our ongoing research projects, which is analyzing the potential of voice monitoring in congestive heart failure.
I would like to thank especially Fan Wu, my colleagues (@DavidCleres and E…https://t.co/ycY506HMdZ
📢 Open for submissions 📢
A research topic led by Tobias Kowatsch, Varun Mishra, Filipe Barata, Felix Wortmann, Katarzyna Wac and Elgar Fleisch
📅Closing date: 16th May 2023
Get involved ➡️ https://t.co/3s43PVjsSo
About six years ago, I embarked on a mission to enable smartphones to detect coughs. Today, I am thrilled to announce that our research has yielded more evidence of how this technology can be integrated into clinical practice. In c…https://t.co/K0vAasnl0r https://t.co/QgQ83Uiwyo
My first lecture at ETH has come to an end! It was very motivating to work with students from different backgrounds and to have the opportunity to raise interest in the field of digital biomarkers! I would like to thank all the students, startups, and tea…https://t.co/5gmqQlCiLM
• If you're wrong, admit it.
• If you're confused, ask questions.
• If you're stuck, seek for help.
• If you make mistake, learn from it.
• If you learn something, teach others
Worse #AI can ➡️ better human decisions.
Our new paper shows this in 1000s of human-AI experiments: making AI's output to users overconfident can improve human decisions > using calibrated AI.
We also propose framework for training AI to opt human perf https://t.co/wfNYdCZtnZ
Lecun (@ylecun)’s 2022 paper on Autonomous Machine Intelligence rehashes but doesn’t cite essential work of 1990-2015. We’ve already published his “main original contributions:” learning subgoals, predictable abstract representations, multiple time scales…https://t.co/Mm4mtHq5CY
New publication by Stefan Jokic, David Cleres, Frank Rassouli, Claudia Steurer-Stey, Milo Puhan, Martin Brutsche, Elgar Fleisch, and Filipe Barata on a novel algorithm to determine to whom a cough belongs.
Read the publication:
https://t.co/IMXoi1TjyK
#machinelearning#asthma
Transformers can achieve few-shot learning (FSL) without being explicitly trained for it.
New research shows that FSL emerges only when the *training data* is distributed in particular ways that are also observed in natural domains like language.
https://t.co/avnNb2VURw 1/
I am honored to serve as a keynote and panel speaker at ETH Analytics Club x ETH AI Center Conversational AI Event at Data Day 2022 this afternoon!
Join us to have an interesting conversation on Conversational AI and its application to healthcare!
#a…https://t.co/x6beBteFwy
Coursera had started with my machine learning class. I’m delighted to announce that I’ll soon be teaching a new Machine Learning Specialization, created by @StanfordOnline & @DeepLearningAI_. Please help me spread the word! You can sign up here: https://t.co/8WJOnNGz7W