🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

“This year is the last time we have, if we are not to see a carbon rebound,” said Fatih Birol, executive director of the International Energy Agency. World has six months to avert climate crisis, says energy expert theguardian.com/environment/20…

Excited to present our #ICML2020 paper “Interpretations are useful: penalizing explanations to align neural networks with prior knowledge! Come to our session on Wednesday 10pm or Thursday 10 AM (CEST) icml.cc/Conferences/20…

Explanations are vulnerable to adversarial attacks but they don’t have to be! “A simple defense against adversarial attacks on heatmap explanation” presented at the Workshop on Interpretability @ #ICML2020 tomorrow. Work with Lars Kai Hansen arxiv.org/abs/2007.06381

Excited to present "Client Adaptation improves Federated Learning with Simulated Non-IID Clients" at the Federated Learning workshop at #ICML2020 on Saturday 3:35 pm (CEST). Work together w/ Rasmus Malik Høegh Lindrup and Lars Kai Hansen arxiv.org/abs/2007.04806

"A vegan diet is probably the single biggest way to reduce your impact on planet Earth, not just greenhouse gases, but global acidification, eutrophication, land use and water use,..., far bigger than cutting down on your flights or buying an electric car" theguardian.com/environment/20…

Our new paper out now! TLDR: Deep learning systems to detect COVID-19 from chest radiographs rely on confounding factors rather than medical pathology. I also wanted to make a quick thread to highlight a few of our findings w/ broader implications: medrxiv.org/content/10.110…

Yesterday I successfully defended my PhD on explainability in neural networks! So happy about the past three years and looking forward to the next chapter. Thanks to my fantastic supervisor Lars Kai Hansen , my insightful committee as well as friends and family for their support

👏👏👏 Danish AI Center is in the last phase of negotations dg.dk/en/fem-fonde-v… DTU DTU Compute

Today at 14:00 CET, Laura Hannemose Rieger and Umang Bhatt will be talking about explainability in machine learning at our seminar series. Don't hesitate to join if you are interested ! cogsystalks.github.io/event-7/

What an amazing day BIG-MAP! Your biannual meeting attracted nearly 100 talented and hardworking people across Europe in person in addition to those who participated online. #Battery #Research #Innovation #Data #Platform #EU Tejs Vegge k

Our work on predicting battery lifetime based on initial cycles with Elixabete Ayerbe Ole Winther Tejs Vegge Arghya Bhowmik and others in BIG-MAP is out!

Same story applies - the unprecedented loss of #Antarctic sea ice for this time of year... More graphs at zacklabe.com/antarctic-sea-…. Data from National Snow and Ice Data Center.

This new infographic presents work from Rieger et al. (Laura Hannemose Rieger, Elixabete Ayerbe, Ole Winther, Tejs Vegge, Arghya Bhowmik; DTU Energy) to efficiently predict battery degradation. Find out more in their #openaccess article: doi.org/10.1039/D2DD00… BIG-MAP, BATTERY 2030 +

Congratulations the BIG-MAP Postdoc Awardees: Eibar Flores SINTEF, Laura Hannemose Rieger DTU Energy, Quentin Jacquet CTA for their stellar work on enhancing the scientific objectives and taking a leading role in establishing a collaborative mindset in BIG-MAP🙌 BATTERY 2030 +

🚀 1D neural network decodes organic molecules' infrared spectra with high accuracy. Unveiling classic group frequencies and non-intuitive patterns, our #XAI method aligns with experts' findings. With M Wilson Tejs Vegge E Flores. shorturl.at/uGKP0 BIG-MAP BATTERY 2030 +

Interested in non-invasive in-situ battery degradation analysis with P2D models? Read more in the new paper by @willappag Laura Hannemose Rieger Arghya Bhowmik DTU Energy and Eibar Flores SINTEF BIG-MAP BATTERY 2030 + authors.elsevier.com/sd/article/S23…

Our PerQueue manager to dynamically orchestrate AI-accelerated modelling and experimental workflows is finally published. Great work Benjamin William Armando Martin Laura Hannemose Rieger Tejs Vegge Juan María García Lastra DTU Energy BIG-MAP BATTERY 2030 + Digital Discovery pubs.rsc.org/en/content/art…

Our paper on using active learning to speed up segmentation of electrode microstructures is out! BIG-MAP BATTERY 2030 + Arghya Bhowmik DTU Energy sandrine lyonnard sciencedirect.com/science/articl…

Our open-source benchmark dataset and code for phase field simulations is live! Designed to accelerate ML development for microstructure evolution using physics-based modeling with U-Nets. Check it out: nature.com/articles/s4159… Arghya Bhowmik BIG-MAP BATTERY 2030 +