The second research symposium with ADA@Aachen and ADA@LIACS members at the magnificent Oude Sterrenwacht in Leiden. So many inspiring talks by all group members.
Very happy to be here today at the the European Parliament, together with @rwth Professor Hermann Ney, one of the worlds leading experts on AI-based language technology (large language models, …) and AI experts from around Europe.
Getting ready for my statement …
Our workshop "Neuro-Explicit AI and Expert-informed Machine Learning for Engineering and Physical Sciences", co-located with
@ECMLPKDD
2023 will take place in Turin, Italy! Submission deadline 12 June. https://t.co/JyMudThx2b
🎉 Our paper on neural network verification won the best paper award at the #AAAI 2023 Workshop on Safe AI:
https://t.co/2kMstqFWdH
Here’s the paper:
https://t.co/QwteFbV1bx
It was great working on this with Matthias König, Annelot Bosman, Jan van Rijn @LIACS!
It's great to see CLAIRE connecting so strongly with one of Europe's most competitive and prestigious programmes for supporting excellence in research and innovation!
I am immensely proud of my former student, Yasha Pushak, who’s been formally awarded his PhD at @UBC earlier today. Working closely with Yasha (here shown in his PhD regalia, with his baby daughter) over the past few years has been very enjoyable and highly rewarding!
Developing #NeuralArchitectureSearch for #EO classification is important to get the best results. We write about our work on this topic on the @ada_research blog: https://t.co/pt0F8kizfF. This includes an advanced #AutoKeras tutorial on the topic!
Last week in Prague TAILOR scientists spent two days in programming the future of #TrustworthyAI for #Europe.
Find their impressions shared on social media here ⤵️
https://t.co/j3Pyq2KgXj
We recently released a new version of Sparkle with support for parallel algorithm portfolios, enabling you to utilise your parallel processing power and get the best performance out of your algorithms. Find Sparkle and examples from our repository at: https://t.co/eZCjbLvEMl
In one sentence: Sparkle aims to make meta-algorithms, such as automated algorithm configuration and selection, accessible and to facilitate their easy and correct use. It is developed and maintained by the ADA research group @ada_research
What does AutoML need for creating accurate COVID-19 forecasting models? Jaco Tetterroo, @HolgerHoos and I studied global open mobility and mortality datasets and showed how concept drift adaptation strategies can help creating forecasting models for this challenging problem.