Fabrizio Ventola

PhD candidate at AI & ML lab @CS_TUDarmstadt (he/him). Research on deep, flexible, and efficient probabilistic models. Mastodon:

[Fabrizio Ventola @fabri_ven]

I'm humbled and happy for the runner up award. Glad that our work on making HPO and, more generally, BO more interactive with Probabilistic Circuits (models providing efficient & flexible inference) has been so well received! Big thanks to my co-authors & to AutoML organizers!

[AutoML Conference @automl_conf]

Best paper award!! Congrats to the authors, super deserved! 🏆

Check it out here: https://t.co/qefUUuQzqV

#AutoML25 #AutoMLconf

[Fabrizio Ventola @fabri_ven]

Happy that our paper on how to make HPO & NAS (& BO) more interactive via Probabilistic Circuits has been well-received! We leverage PCs' flexible inference to incorporate human feedback on arbitrary subsets of hyperparameters anytime during optimization! Check it out 👇

[Jonas Seng @jonas_seng]

🚀 Do you want to nudge your Bayesian Optimization into the right direction effectively? Then check out our new work Hyperparameter Optimization via Interacting with Probabilistic Circuits which got accepted at this year's @automl_conf!

[antonio vergari ⚔️ @tetraduzione]

the #TPM ⚡Tractable Probabilistic Modeling ⚡Workshop is back at @UncertaintyInAI #UAI2025!

Submit your works on:
- fast and #reliable inference
- #circuits and #tensor #networks
- normalizing #flows
- scaling #NeSy #AI

🕓 deadline: 23/05/25
👉 https://t.co/qj0llWEMdU

[Pasquale Minervini @PMinervini]

Folks, how would you go about this?

[antonio vergari ⚔️ @tetraduzione]

Less than two weeks to submit your papers on:

📈 #lowrank adapters and #factorizations
🧊 #tensor networks
🔌 probabilistic #circuits
🎓 #theory of factorizations

to the first workshop on connecting them in #AI #ML at @RealAAAI

please share! ���

👇👇👇
https://t.co/m65ScBfTyL

[clem 🤗 @ClementDelangue]

"You can't chatbot your way out of climate change". Great points from @SashaMTL from @huggingface on @BBC!

[Antonia Wüst @toniwuest]

Our recent workshop paper puts Vision-Language Models like #GPT4o to the test with Bongard problems🧩
We show that VLMs struggle to identify concepts that are quite intuitive to humans. Still a long way to go for human-like visual reasoning! 🤖🧠
arXiv: https://t.co/JKdAXvJwfA

[Kristian Kersting @kerstingAIML]

Aktuelle #KI-Modelle bestehen nicht KI-Benchmarks aus den 1960iger Jahren 😤 Tolle Zusammenarbeit mit @toniwuest @philosotim @lukas_helff @devendratweetin @c_rothkopf

https://t.co/nddarhIyuc

[Kristian Kersting @kerstingAIML]

🙏for the kind words. I am standing on the shoulders of giants 🦾 xLSTMs are amazing and work really well for multivariate time series. All the credit to the amazing team @mkraus_io @felixdivo @devendratweetin

[Lorenzo Loconte @loreloc_]

We learn more expressive mixture models that can subtract probability density by squaring them

🚨We show squaring can reduce expressiveness

To tackle this we build sum of squares circuits🆘

🚀We explain why complex parameters help, and show an expressiveness hierarchy around🆘

[Fabrizio Ventola @fabri_ven]

@automl_conf Joint work w/ @jonas_seng @fabian_kalter @an_der_Modau @kerstingAIML

[Fabrizio Ventola @fabri_ven]

Back from @automl_conf, truly enjoyed it!🙂Great friendly community. Got constructive comments & positive feedback. Glad our works on #probabilistic #circuits have been so well received! Big thanks🙏 to the organizers. Stay tuned for our next developments on PCs in NAS & HPO. 👇

[Jonas Seng @jonas_seng]

Ever wondered how to efficiently build hybrid architectures for accurate probabilistic time series forecasting? Then check out our recent @automl_conf paper!

Paper: https://t.co/3AyzncPGtB

Joint work with @fabian_kalter, @an_der_Modau, @fabri_ven and @kerstingAIML.

Details ⬇️

[Fabrizio Ventola @fabri_ven]

Please consider submitting your works on #Tractable #Probabilistic #Models at the #TPM2024 workshop @ #UAI2024 @UncertaintyInAI --- still ~10 days left

👉 https://t.co/4EZbyToJvR

🗓️Deadline: May 27th, 2024 AoE

#probabilistic #circuits #normalizing #flows #pgm #trustworthy #AI

[Martin Mundt @mundt_martin]

We’ve also shown at UAI23 w/ @fabri_ven @sbraunmz @kerstingAIML that overconfidence isn’t unique to NNs, but also practically occurs in probabilistic circuits - tractable models.

We’ve shown how we can combat this through tractable uncertainty in PCs: https://t.co/9hvm2XsNov

[Kristian Kersting @kerstingAIML]

Happy and proud to have @MIT's Tomaso Poggio @TomasoPoggio give a talk at @CentreForCogSci and @Hessian_AI ! @TUDarmstadt

[Zhongjie @an_der_Modau]

Thanks to everyone who joined our oral 1C session or discussed later at the poster for Characteristic Circuits at #NeurIPS2023 Check out our paper at: https://t.co/wYDFaAfXzF @kerstingAIML

[Lorenzo Loconte @loreloc_]

We convert popular knowledge graph embedding (KGE) models such as ComplEx into generative models of triples that

1️⃣ better scale training to very large knowledge graphs
2️⃣ make reliable predictions with logical constraints
3️⃣ support sampling

in our #NeurIPS2023 oral!

[Fabrizio Ventola @fabri_ven]

😂🔝@MBrack_AIML

[Yann LeCun @ylecun]

Okay, this has got to be the absolute best use of text-to-image technology ever 😎😂😂😂

[Kristian Kersting @kerstingAIML]

This will be an exciting @RealAAAI 2024 workshop: LLMs and causality? Are they causal 🦜? What does that imply for reasoning and planning? Looking forward

[Kristian Kersting @kerstingAIML]

📢Open-source #AI has a lot of pros ... but also cons. as discussed in our recent preprint

👉Balancing Transparency and Risk: The Security and Privacy Risks of Open-Source Machine Learning Models
👉https://t.co/K12ujGEoHP

Still a lot work to be done,

https://t.co/aKpUsMmNUi