Jonathan Wenger
@JonathanWenger5
Followers
234
Following
500
Media
17
Statuses
104
Postdoc @Columbia. PhD @uni_tue with @PhilippHennig5. I enjoy cycling.
Joined May 2018
RT @EurIPSConf: EurIPS is coming! 📣 Mark your calendar for Dec. 2-7, 2025 in Copenhagen 📅. EurIPS is a community-organized conference where….
0
25
0
We have extended the deadline to May 26th, so fear not if your NeurIPS submission still needs work. There is still time to submit to the CODEML workshop @icmlconf 2025!.
Built a new ML library? Maintain a crucial project? Improved OSS practices? Your work deserves recognition! Submit your contributions to the CODEML workshop @ #ICML2025. We're championing open-source in ML. 💻✨ Deadline May 19.
0
0
0
We have a fantastic lineup of speakers who have made deep contributions to open-source in ML, e.g. @sarahookr, @ChrisRackauckas, @SingularMattrix, @tri_dao, @BlancheMinerva, and Evan Shelhamer!.
Built a new ML library? Maintain a crucial project? Improved OSS practices? Your work deserves recognition! Submit your contributions to the CODEML workshop @ #ICML2025. We're championing open-source in ML. 💻✨ Deadline May 19.
0
5
16
Built a new ML library? Maintain a crucial project? Improved OSS practices? Your work deserves recognition! Submit your contributions to the CODEML workshop @ #ICML2025. We're championing open-source in ML. 💻✨ Deadline May 19.
1
4
10
RT @MPI_IS: Researching #AI #MachineLearning #Robotics or #HCI? Join our elite #PhD program - a partnership with @uni_stuttgart & @uni_tue!….
0
39
0
RT @AJLUnited: The critically acclaimed @codedbiasmovie is now on @Netflix. Follow AJL founder @jovialjoy as she discovers that facial rec….
0
26
0
Come chat with us at our poster #313 today at 4pm!.
Did you know you can perform ❗exact❗ uncertainty quantification for approximate Gaussian processes in quadratic time? 🤯. Joint work w/ G. Pleiss, M. Pförtner, @PhilippHennig5 and J. Cunningham at #NeurIPS2022. Paper 📄 Code 💻
0
0
2
I’ll be at @NeurIPSConf in New Orleans all week! DM me if you want to grab a coffee and chat about #gaussianprocesses, #numericsforml, #probabilisticnumerics, or anything else!.
0
0
0
Different ways to approximate, i.e. target computation, recover and improve existing methods (Cholesky, CG, . ) with at most quadratic cost!. The method is fully online, connects to generalized GP regression, and more. Find out more and chat with us @NeurIPSConf next week!
0
0
2
Doing so, allows us to exactly quantify the relative error to the true latent function, precisely in the same way a GP does, but at the cost of an approximate method!
1
0
0
In the same way, that limited data induces uncertainty about the true function, so does limited computation! We quantify computational uncertainty using techniques from #probabilisticnumerics to improve uncertainty quantification.
1
0
1
Approximations allow us to scale GPs to large datasets, but usually uncertainty quantification suffers as a result. But, why does this happen?
1
0
1
Did you know you can perform ❗exact❗ uncertainty quantification for approximate Gaussian processes in quadratic time? 🤯. Joint work w/ G. Pleiss, M. Pförtner, @PhilippHennig5 and J. Cunningham at #NeurIPS2022. Paper 📄 Code 💻
2
5
39
Come to beautiful Tübingen and learn more about probabilistic numerical methods and uncertainty quantification!.
📢 Are you keen on learning more about probabilistic numerics, scientific computing, uncertainty quantification, and probabilistic inference? 📢. Then sign up for the *first ever* #ProbNumSchool:. March 27-29, 2023, Tübingen. @PhilippHennig5 (thread 👇)
0
0
3
RT @MPI_IS: Researching #AI #MachineLearning #Robotics #ComputerVision? Join our elite #PhD program - a partnership with @uni_stuttgart & @….
0
85
0
RT @haesleinhuepf: 🚨Game changer alert🚨 for scientists in Germany who publish and maintain #opensource software and #opendata. Our @github….
0
273
0
RT @fx_briol: Some colleagues and I are organising a workshop on score-based methods at NeurIPS 2022! The deadline for workshop papers is t….
0
6
0
If you are at ICML come chat with us at the poster session tonight or see the oral at 2pm in room 307.
Did you know you can use preconditioning not only for bias, but also variance reduction💡 in Gaussian process hyperparameter optimization?. Joint w/ G Pleiss, @PhilippHennig5, J Cunningham and J Gardner at #ICML2022. Paper 📃 Code 💻 Part of #gpytorch
0
0
3