Physics MSc. PhD student @UvA_Amsterdam. Deep Learning 🧠 & Dynamical Systems 💫 Physical AI 🤖
Amsterdam, Holland
Joined September 2019
- Tweets101
- Following633
- Followers285
- Likes1.4K
Pinned Tweet
Group equivariance + NeuralODEs + Attention-based Neural Fields + MAML = sota PDE forecasting! 🇨🇦
🇨🇦 Deeeelighted to share that this work got into #neurips2024. Many thanks to my dear friend and co-author @Dafidofff, as well as the rest of the team.
Solving PDEs in continuous space-time with Neural Fields on cool geometries while respecting their inherent symmetries! 💫💫
Riccardo Valperga retweeted
If you want to understand “Generative Modeling via Drifting” you can read
1. Ai-Sampler arxiv.org/abs/2406.02490 by me and @RValperga
2. Particle Dynamics for Learning EBMs arxiv.org/abs/2111.13772 by @k_neklyudov
Riccardo Valperga retweeted
Grounding Continuous Representations in Geometry: Equivariant Neural Fields
by @Dafidofff @davidmknigge @samuelepapa_ @RValperga @SharvVadgama @egavves @erikjbekkers
➡️ openreview.net/forum?id=A4eC…
⚠️ Hall 3 + Hall 2B #558
xitter.cf/Dafidofff/status/18879…
🚀New paper at #ICLR2025!🚀
We introduce Equivariant Neural Fields (ENFs): a new class of Conditional Neural Fields (CNFs) that explicitly ground continuous representations in geometry across image & 3D shapes reconstrution, PDE forecasting & more
🧵Read on for details! 👇
Riccardo Valperga retweeted
🚀New paper at #ICLR2025!🚀
We introduce Equivariant Neural Fields (ENFs): a new class of Conditional Neural Fields (CNFs) that explicitly ground continuous representations in geometry across image & 3D shapes reconstrution, PDE forecasting & more
🧵Read on for details! 👇
Riccardo Valperga retweeted
🔥 3/4 #ICLR2025 Grounding Continuous Representations in Geometry: Equivariant Neural Fields (ENF), for geometry-informed continuous signal representations. Learn more: openreview.net/forum?id=A4eC…
with D. Wessels, @davidmknigge @samuelepapa_ @RValperga @SharvVadgama @erikjbekkers
Ironico. Pur non avendo la più vaga alba di come funzioni un'AI ti lanci in questo genere di commenti. Una persona intelligente, che sa cos'è un LLM non lo antropomorfizzerebbe aspettandosi che questo risponda "non lo so", confrontandolo con una persona cosiddetta "intelligente".
I will be in Montréal until December for my internship with ServiceNow. I will be working on causal discovery from time-series. Get in touch if you are around and want to chat about some **apprentissage profond**
Riccardo Valperga retweeted
It's time for some @GRaM_org_ workshop updates happening at #ICML2024 on the 27th of July.
We have a great list of invited speakers and panelists.
We have Rose Yu @yuqirose , Phillip Isola @phillip_isola , Nina Mialone @ninamiolane , Joey Bose @bose_joey , Zahra Kadkhodaie @ZKadkhodaie , Rianne van der Berg @vdbergrianne and Max Welling @wellingmax .
Learning to sample: time-reversibilty meets Metropolis-Hastings. Check our #icml2024 paper where we propose an objective that upper-bounds the TV distance between the stationary distribution of a parametric Markov kernel, and the distribution you want to sample from! (1/6)
By bootstrapping on its own samples, the kernel gets better at sampling. This means better exploration and effective sample size. (5/6)
Check the paper by @eeevgen, myself and @egavves at arxiv.org/abs/2406.02490. Come to our poster Tuesday morning (Hall C 4-9 #1708). Code available at github.com/Ricvalp/aisampler. (6/6)
Check our group equivariant PDE surrogate based on conditioned neural fields. Better than existing operator-based methods when it comes to sparse and irregularly sampled data!
Riccardo Valperga retweeted
Fantastic to see this innovative work by @Dafidofff and @davidmknigge on Equivariant Neural Fields. Creative and refreshing perspective on equivariant representation learning and neural fields. Worth a read and follow!
🥠 Excited to introduce our latest work on Equivariant Neural Fields (ENFs)! Grounding conditioning variables in geometry 🚀
Paper: arxiv.org/abs/2406.05753
Github: github.com/dafidofff/enf-jax
Project Page: dafidofff.github.io/enf-jax/
Details below 👇👇
Cross-attention strikes again👀.
🥠 Excited to introduce our latest work on Equivariant Neural Fields (ENFs)! Grounding conditioning variables in geometry 🚀
Paper: arxiv.org/abs/2406.05753
Github: github.com/dafidofff/enf-jax
Project Page: dafidofff.github.io/enf-jax/
Details below 👇👇
Riccardo Valperga retweeted
I will be presenting the poster at Arch 4A-E Poster #297.
Stop by to know more about using implicit neural representations!
🎉 Our paper "How to Train Neural Field Representations" (fit-a-nef.github.io/) has been accepted at #CVPR2024! Grateful to the team @RValperga @davidmknigge @MiltosKofinas @phillip_lippe and supervisors @efstratios
and Jan-Jakob Sonke!🙌 (1/4)
Riccardo Valperga retweeted
Grounding Continuous Representations in Geometry: Equivariant Neural Fields
@Dafidofff @davidmknigge , @samuelepapa_, @RValperga, @SharvVadgama, @erikjbekkers @egavves
tl;dr: capsules done right? Or how to merge a car and a duck :)
arxiv.org/abs/2406.05753
