In case you participate at AAAI 2023:
Today, I will present our paper on "Boundary Graph Neural Networks for 3D Simulations" (https://t.co/dqMQkWiock) in room Room 146B at 2:00 pm.
In the evening, I will also be available at our poster.
I am happy to share that I am joining Pumas-AI as a Machine Learning Scientist! I look forward to learning from the great team at Pumas-AI, and to contributing to their effort to bridge science, healthcare, and machine learning. @pumas_ai
After all these very enriching years, first at the Institute of Computational Perception, and then at the Institute for Machine Learning, it's time to leave Johannes Kepler University Linz. A heartfelt thank you to all of my colleagues. All the best!
@cpjku@LITAILab@jkulinz
Excited to share our work on history compression via language models in RL, presented at #ICML2022🤩🤩. Our novel framework HELM⎈ augments an agent with a history compression module which leverages a pretrained language Transformer without any training or finetuning 🤯🤯
1/5
ELLIS "Machine Learning for Molecules" workshop tomorrow! https://t.co/01DnDo41zr Hosted by the ELLIS unit Cambridge and ELLIS unit Linz. Speakers from pharma and academia on new ML techniques for molecules.
On December 13 I will present our work “BioassayCLR: Prediction of biological activity for novel bioassays based on rich textual descriptions” at the ELLIS ML for Molecule Discovery WS. Joint work with @HochreiterSepp and @gklambauer. https://t.co/0OyNPpfv4C #ML#DrugDiscovery
@ArsElectronica Not the robot dog. It normalizes a military & police weapon. It was uncomfortable to see it at the festival carrying prizes and interacting with kids. Please uplift works of criticism, action and joy instead!
Can musical genres be treated as distinct data domains in the context of ML? We show that domain adaptation can help generalize a perceptual feature model to a specific genre not present in the training set. #ICASSP2021@ieeeICASSP
https://t.co/1va7eOJ7Nz
Our article "The Promise of AI for DILI Prediction" has been published at @FrontAIBigData. We provide an up-to-date review of AI approaches to drug-induced liver injury prediction and discuss current challenges and future directions. https://t.co/3CH9bO4WrV #ML#DrugDiscovery
We introduce Align-RUDDER, which enables Reinforcement Learning from few demonstrations by
reward redistribution via multiple sequence alignment.
Paper: https://t.co/nos4JxAWuZ
Blog post, including a demonstration video of mining a diamond in Minecraft: https://t.co/xa8WccPBHL
Self-attention mechanism can be viewed as the update rule of a Hopfield network with continuous states.
Deep learning models can take advantage of Hopfield networks as a powerful concept comprising pooling, memory, and attention.
https://t.co/FL8PimjVo9
https://t.co/HT79M95lkn
Thanks for the detailed and approachable explanation of our paper, @musikalkemist! Looking forward to more of your paper breakdowns. Much needed in the #mir and #aimusic community.
@AndreuVall@veroamilbe
https://t.co/EaYUyC1hE7
Crema una de les comissaries de Minneapolis després de l'assassinat policial de #GeorgeFloyd i la repressió que ha colpejat les manifestacions posteriors.
#BlackLivesMatter#minneapolisriots
En el 75 aniversario de la liberación del campo de concentración de Mauthausen y subcampos, la @amauthausen no ha podido desplazarse hasta Austria pero la memoria por todas las víctimas está presente hoy en Linz.
#75Befreiung#75Liberación