Dominik Fuchsgruber
@dfuchsgruber
Followers
55
Following
28
Media
5
Statuses
37
PhD Student @TU_Muenchen | Uncertainty Estimation, Spatio-Temporal Learning
Munich
Joined June 2024
Excited to announce our #ICLR2025 spotlight work deriving the first exact certificates for neural networks against label poisoning 🎉. Joint work w/ @maha81193, @guennemann & Debarghya. For more details check out the thread below👇 or check out our paper https://t.co/MPX1EqDCPG.
arxiv.org
Machine learning models are highly vulnerable to label flipping, i.e., the adversarial modification (poisoning) of training labels to compromise performance. Thus, deriving robustness certificates...
🎉Excited to announce our #ICLR2025 Spotlight! 🚀@lukgosch and I will be presenting our paper on the first exact certificate against label poisoning for neural nets and graph neural nets. Joint work with @guennemann and Debarghya 👇[1/6]
0
15
23
Happy to share that our paper "Lift Your Molecules: Molecular Graph Generation in Latent Euclidean Space" got accepted to #ICLR2025, and we will be presenting it this week in Singapore! Joint work with @n_gao96, @TomWollschlager, @j_m_sommer, and @guennemann. 🧵 1/
1
18
29
I am truly excited to share our latest work with @MScherbela, @GrohsPhilipp, and @guennemann on "Accurate Ab-initio Neural-network Solutions to Large-Scale Electronic Structure Problems"! https://t.co/By8CetKwHU
arxiv.org
We present finite-range embeddings (FiRE), a novel wave function ansatz for accurate large-scale ab-initio electronic structure calculations. Compared to contemporary neural-network wave...
2
32
108
The paper of my final "pure" physics project just got published in @NanoLetters
https://t.co/vf1AciNaru joint work with Ludwig Burger, @ceslopast, Hamid Seyed-Allaei @GiovanniGiunt20 and Ulrich Gerland
0
9
14
Do you think your LLM is robust?⚠️With current adversarial attacks it is hard to find out since they optimize the wrong thing! We fix this with our adaptive, semantic, and distributional objective. By @guennemann's lab & @GoogleAI, w/ @ai_risks support Here's how we did it. 🧵
2
14
20
Safe and reliable machine learning has never been more relevant. However, at the same time, LLMs made robustness research **more complex, less reproducible, and harder-to-evaluate**. How can we enable research progress despite these challenges?🧵 @guennemann @gauthier_gidel
1
20
44
Congrats to my amazing PhD students: We have 9 papers accepted at #ICLR2025. Reliability, AI4Science, graphs, LLMs, and more ( https://t.co/O3f86agVaN). And if you follow the recent discussions about AI efficiency, you might like our blog and webinars ( https://t.co/MsohgAJRUZ).
4
14
179
Super happy & honored that our work on certifying NNs against poisoning won the Best Paper Award @AdvMLFrontiers at #NeurIPS2024. Come by our poster 10:40am-12&4-5pm (or talk) tomorrow :) Joint work w/@maha81193, Debarghya Ghoshdastidar & @guennemann L: https://t.co/PrZi2nlUtY
2
15
30
Make sure to stop by our #NeurIPS poster on Spatio-Spectral Graph Neural Networks (S²GNNs)! An efficient synergy of spatially & spectrally parametrised graph convolutions. Joint work w/ @geisler_si @dan1elherbst @guennemann. 📎 https://t.co/wzX5M67oPY 📆 Dec 13, 11am, #4608
0
14
24
🏃♂️Our model can applied both post-hoc at negligible cost to many popular GNNs! 🔦We also establish a connection to evidential learning that drastically improves the robustness to distribution shifts in terms of accuracy.
1
0
5
👀We use a latent space density model for regularization that provably solves overconfidence issues far from the training data. 🔝Our uncertainty estimate is the only one that consistently performs well over an extensive family of distribution shifts.
1
0
4
We use Energy-based models to quantify epistemic uncertainty in GNNs! ⚡️We compute energy entirely agnostic of the graph and then interleave graph diffusion and energy marginalization to combine uncertainty at different structural levels.
1
0
4
Excited to present our spotlight paper on uncertainty for GNNs at #NeurIPS! 📝Paper: https://t.co/YW8B7tXqoo 📆Come by our poster on Dec 12th at 11am! Thanks to my amazing collaborators @TomWollschlager and @guennemann!
1
9
25
Excited to present our work on Neural Pfaffians at #NeurIPS. 🗣️ Oral: Friday 3:30pm, East Ballroom A, B 📊 Post: Friday 4:30pm - 7:30pm, East Exhibit Hall A-C #3600 📝 Paper: https://t.co/X54t2hhfZs Happy to chat!
0
11
31
This week, we will present our recent #NeurIPS2024 paper. 📎 Paper: https://t.co/92UJBKn4WZ 📆 Make sure to visit our poster #2600 on Fri, 13 December at 11 am! Joint work with my amazing mentors @leon_het @j_m_sommer @fabian_theis @guennemann
1
16
34
Next week, I'll present our recent paper at NeurIPS 2024 in Vancouver. Many thanks to my amazing collaborators @Bertrand_Charp, @DanielZuegner , and @guennemann!
1
12
20
Monday with @n_gao96 in the reading group "Learning Equivariant Non-Local Electron Density Functionals" https://t.co/RfDtZSmO03 Join us on zoom at 9am PT / 12pm ET / 6pm CET: https://t.co/R8d1EHxLCx
5
17
99
Deep learning with differential privacy can protect sensitive information of individuals. But what about groups of multiple users? We answer this question in our #NeurIPS2024 paper https://t.co/PemQWF3PAq Joint work w/ @mihail_sto @ArthurK48147 @guennemann. #Neurips (1/7)
1
12
21
Switching gears from QMC to DFT for this one. I'm excited to share our newest work, where we learn the non-local exchange-correlation functional in KS-DFT with equivariant graph neural networks! Joint work w/@ESEberhard, @guennemann 📝 https://t.co/2Q1Vy87R60
4
27
105