Zhongjie

Give me higher likelihood!

[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!

[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 ⬇️

[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

[Zhongjie @an_der_Modau]

@apratimbh @iclr_conf Maybe because of reviewer names unveiled?🙈

[Steven Braun @sbraunmz]

Interested in how to tackle overconfidence and introduce uncertainty in #ProbabilisticCircuits? 🧠

Join our oral pres. on "Probabilistic Circuits That Know What They Don’t Know" Thu 3:15PM and chat with us at our poster (118) later at 4:30PM @ #UAI2023! 🎉

[Fabrizio Ventola @fabri_ven]

Probabilistic circuits “don’t know what they don’t know”… but we can easily fix it with uncertainty estimation in a tractable analytical way, inspired by Monte Carlo dropout! Work w/ @sbraunmz @an_der_Modau @mundt_martin @kerstingAIML

Paper + code: https://t.co/IiSsCithxh
🧵⬇️

[Fabrizio Ventola @fabri_ven]

Happy to announce the next speaker for the KompAKI seminar series @CS_TUDarmstadt. For December's event, we will have Eduardo dos Reis from @DMTUDA with a talk on DiffML: End-to-end Differentiable ML Pipelines. Looking forward to it! 😃

[Martin Mundt @mundt_martin]

It’s not everyday you see your advisor on national evening TV.

Super cool!

👏🎊🎉 congrats @kerstingAIML

https://t.co/YC7wIX8027

[Zhongjie @an_der_Modau]

this is joint work with Nils, @devendratweetin @mundt_martin and @kerstingAIML 😃 Welcome for a discussion at #UAI2022

[Fabrizio Ventola @fabri_ven]

If you are at #UAI2022 and interested in hybrid deep architectures w/ probabilistic circuits & neural predictors, spectral time series forecasting & uncertainty estimation, join us at today's poster session, @an_der_Modau and I will tell you more about Predictive Whittle Networks

[Zhongjie @an_der_Modau]

First day in Eindhoven at #UAI2022 is already a lot of fun ;-)

[Fabrizio Ventola @fabri_ven]

At @UncertaintyInAI in #Eindhoven, tomorrow we'll present "Predictive Whittle Networks for Time Series" a hybrid architecture for accurate, efficient and trustworthy forecasting w/ @an_der_Modau Nils Thoma @devendratweetin @mundt_martin @kerstingAIML info: https://t.co/bvmoHMAVY4

[Martin Mundt @mundt_martin]

🥳Accepted #UAI: "Predictive Whittle Networks for Time Series": https://t.co/bNdFMk5zZi

TLDR: We synergize powerful spectral neural forecasters & tractable likelihoods from probabilistic circuits -> better predictions + uncertainty

Great work by @an_der_Modau w/ @kerstingAIML

[Zhongjie @an_der_Modau]

Another upsampling from MOMoGP. Check out our paper at #uai2021 https://t.co/fNFmR7ZIzt
See you at the poster session today @martin_trapp @kerstingAIML

[Sterling Crispin 🕊️ @sterlingcrispin]

Upscaling to 1,000x1,000px after being shrunk to 50x50px :

Super Resolution GAN, Bicubic Smoother, Preserve Details (2.0), Nearest Neighbor

[Apratim Bhattacharyya @apratimbh]

Our new #CVPR2021 paper "Euro-PVI: Pedestrian Vehicle Interactions in Dense Urban Centers" introduces a new dataset and a CVAE based method for interaction aware trajectory prediction. 1/n
Paper: https://t.co/bE1VHLyYC2
Code: https://t.co/3IP87J38Hb
Data: https://t.co/ce6qjMX5Mz

[Kristian Kersting @kerstingAIML]

Here the underlying @icmlconf 2021 paper on Whittle Networks 👉 https://t.co/9y2QTkiNZO. Really amazing work by Zhongjie Yu & Fabrizio Ventola. #Probabilistic #Circuits over complex valued random variables to estimate, e.g., the Whittle likelihood of time series.

[Zhongjie @an_der_Modau]

I will take a look at this post 10 years later, just to see what a crazy time it is now

[Zhongjie @an_der_Modau]

@martin_trapp @arnosolin @CSAalto Congrats!

[Ferenc Huszár @fhuszar]

How it started: How it's going:

[Zhongjie @an_der_Modau]

Orz https://t.co/QN5KmQjBV6

[Zhongjie @an_der_Modau]

610104199001104430