antonio vergari ⚔️ not at #ICLR2025
@tetraduzione
Followers
5K
Following
32K
Media
877
Statuses
11K
human being | assoc prof in #ML #AI @ancAtEd | PI of #APRIL | #reliable #probabilistic #models #tractable #generative #neuro #symbolic | heretical empiricist
Joined February 2015
🗓️ Deadline extended: 💥2nd June 2025!💥 . We are looking forward to your works on: . 🔌 #circuits and #tensor #networks 🕸️ .⏳ normalizing #flows 💨 .⚖️ scaling #NeSy #AI 🦕 .🚅 fast and #reliable inference 🔍 . & more! . please share 🙏.
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 .👉
0
7
19
One trick that I like to use when training my neural networks is to add some noise ε~Laplace(time(), sqrt(time())) to the gradients of the 13th layer at epoch 3 for batch 7.
30
65
961
Is there a reason why a BSc, MSc or PhD Student in #AI or #ML should not attend #UAI2021 @UncertaintyInAI on July 27-30?. The conference is top tier, will be virtual and registration will be free for students!!!.👇👇👇.
4
62
243
#Neural nets struggle to #guarantee their predictions conform to our prior knowledge!. We devise a drop-in replacement layer that injects given #symbolic #constraints while retaining #exact and #efficient gradient optimization in our #NeurIPS paper:. 1/🧵
7
40
239
Classical mixture models are limited to positive weights and this requires learning very large mixtures!. Can we learn (deep) mixtures with negative weights?. Answer in our #ICLR2024 spotlight by @loreloc_ Aleks, Martin, Stefan, Nicolas @arnosolin . 📜
After the score-based models, I decided to take a step back and cover the basics in a new blog post: Probabilistic modeling and Mixture Models. Additionally, I make a brief intro to ✨Probabilistic Circuits✨ Check:.📄Post: 🖥️ Code:
5
45
217
I am looking for troublemakers who want to start a PhD in #probabilistic #ML #AI at @ancAtEd @InfAtEd @EdinburghUni to design the foundations for the next generations of #autonomous #systems. Drop me an email or DM here!.Please share!.
We have 14 funded places available on our PhD in Robotics and Autonomous Systems programme starting September 2022. More details and how to apply
7
61
198
What are some must-read papers on diffusion models?. I am looking at a mandatory literature as well as some good intros. cc @jmtomczak.
13
24
198
I have fully-funded PhD positions (3.5 yrs) for troublemakers in #ML #AI who want to design the next gen of #probabilistic #models and #programs that are provably #reliable and #efficient. Join @InfAtEd @EdinburghUni. Email me! ✉️.Please share! 🔁. Apply 👉
4
74
193
I have fully-funded PhD positions for troublemakers in #ML #AI at @ancAtEd @InfAtEd @EdinburghUni who want to design the next gen of #probabilistic models and programs that are provably reliable and efficient. Email me! ✉️.Please share! 🔃. 👉
6
68
178
Today I feel privileged to see my #ERCStG proposal . #UNREAL - A Unified Reasoning Layer for Trustworthy #ML . being funded by @ERC_Research! .I will be working on #probabilistic #reasoning and #programming to #automate #realiable model design & deployment. But first a big. .
26
12
167
Protip to all PhD students struggling to write last minute their @NeurIPSConf #NeurIPS2022 papers:. large language models such as GPT3 ( and OPT-175B ( are now publicly available for you to write your missing sections!.
2
11
134
When conference #proceedings were printed in a single book and submissions did not follow a single #LaTeX template: . behold the papers accepted at #NIPS89 the 2nd Conference on advances in Neural Information Processing Systems!. 🧵👇
1
10
134
💥Breakthrough work for the tractable probabilistic modeling community!💥. 👉 Continuous latent variable models can be turned into expressive tractable discrete mixtures via numerical integration. by @ahc_correia @gengala13 @cassiopc @equaeghe @ropeharz.
3
21
129
I'm extremely happy to share that my application to become an @ELLISforEurope #ELLIS Scholar has been rejected!.🥂🍾🤦♂️.
11
1
125
As I joined the @ELLISforEurope Unit at @EdinburghUni @InfAtEd I'm looking for PhD students and other #ELLIS comrades for co-supervision!. Topics: efficient and reliable generative models in the wild; automating learning, reasoning and their combination!. Please share 🔁 ❤️ 🙏 !.
Are you considering a PhD in #machinelearning or a related research area? The next round of applications for the ELLIS PhD program is coming up. Stay tuned at and apply by November 15!. LINK to call:
3
45
120
Soon the rebuttal phase for @NeurIPSConf #NeurIPS2020 will start . On a Friday (evening in Europe). It is not the first time #ML #AI authors are asked to work on weekends. This is a bad cultural and working habit that can only go worse in times of #COVID19. When for a change?.
3
10
120
TIL there are 3-10% of training images of CIFAR-10/0 that are identical or very similar to test set images. Replacing them with new samples (from same distribution) makes the test set accuracy drop by 9-14%.
Often we take our image datasets for granted. There’s value in actually looking at the pixels. Check out:. Do We Train on Test Data? Purging CIFAR of Near-Duplicates, Björn Barz, Joachim Denzler. Paper:
2
23
122
A deeply interesting tutorial by @fchollet @MelMitchell1 @ChrSzegedy at #NeurIPS2020 . "#Abstraction & #Reasoning in #AI systems: Modern Perspectives ". or "What are abstraction, generalization and analogic reasoning?". 1/.👇👇👇.📺
1
19
119
@skoularidou @_sam_sinha_ @zacharylipton but that uses adversarial learning in one variant (and MMD in another). Let me shamelessly plug RAEs The catch is that ex-post density estimation on the latent space gets you better aggregate posterior estimation and hence sample quality.
4
14
118
great opportunity to do a #PhD in #Europe in #ML #AI. 🚨🚨🚨.I'll hire 2 students via #ELLIS this year, reach out if you want to do research in:. - #reliable and #efficient ML in the wild.- scalable #neurosymbolic #nesy AI.- #lowrank representations.- #tractable inference. 🚨🚨🚨.
The #ELLISPhD application portal is now open! Apply to top #AI labs & supervisors in Europe with a single application, and choose from different areas & tracks. The call for applications: Deadline: 15 November 2024. #PhD #PhDProgram #MachineLearning #ML
2
27
119
2
25
117
A formal definition of what an inductive bias is in #ML from @tommmitchell book (1996). and yes, it can be any sort of background knowledge or prior (we can express in logical rules about the input, and mod at least for Tom's version space)
6
21
110
I met a kid today at the beach who was struggling to make a castle with pebbles. After the nth failure, I taught him the bitter lesson. I told him that scaling is all he needs, that he should not waste time reasoning and that symbols are not well defined. He burst into tears.
5
2
111
Q: Can deterministic #autoencoders (AEs) learn latent spaces that are as smooth and meaningful as #VAEs do?. A: Yes, see gif and read small thread 👇 👇 👇. Work with @ParthaG64920039 Mehdi M.S. Sajjadi @Michael_J_Black @bschoelkopf accepted at @iclr_conf #iclr2020. 👇 1/
3
36
103
I will be hiring through #ELLIS this year: 3 fully-funded PhD positions for troublemakers in #ML #AI who want to design the next gen of #probabilistic #models and #programs that are provably #reliable and #efficient. Join @InfAtEd. Email me! Please share!.
The portal is open: Our #ELLISPhD Program is now accepting applications! Apply by November 15 to work with leading #AI labs across Europe and choose your advisors among 200 top #machinelearning researchers! . #JoinELLISforEurope #PhD #PhDProgram #ML
3
30
96
📜 """.Inconsistent results based on hyperparameter optimization configurations are widespread in #ML. When comparing two algorithms J and K searching one subspace can yield that J outperforms K, but searching another can entail the opposite. """.
4
12
99
Dear reviewers,. if a method is very simple and yet yields big advantages (time/space efficiency, sota performance, . ) and it is novel (no one tried it even if it is simple), it is TOTALLY worth a publication. (I'm talking in general, here).
5
15
96
#NeurIPS2024 writing tips.(part 1). ✨make a very clear example of what is the issue/problem you are trying to solve✨. 👉what is obvious for you is not for others. 👉even experts in the field need to "see it". 👉this can be your Fig 1. 👉it helps scope your solution.
3
5
95
The list of accepted papers at @UncertaintyInAI looks super interesting: 👇. (lots of tractable models as well!).
1
14
87
#ELLIS PhD Applications are now open!. Consider applying to my group at @EdinburghUni @InfAtEd if you want to work in probabilistic ML investigating efficient and reliable models, combining learning and reasoning and automating them!.
ELLIS PhD Program in #machinelearning now accepting applications (deadline: November 15, 2021). Details on the program and the application process here:
2
20
85
from the @NeurIPSConf guide to ACs:. """.Try to counter biases you perceive in the reviews. Unfashionable subjects should be treated fairly but often aren't, to the advantage of the papers on more mainstream approaches. To help the NeurIPS community move faster [1/2].
1
7
87
many of the recent successes in #AI #ML are due to #structured low-rank representations!. but. What's the connection between #lowrank adapters, #tensor networks, #polynomials and #circuits?. join our @RealAAAI workshop to know the answer!. 👇👇👇.
3
16
80
Thanks @dagstuhl for giving us the opportunity to connect communities and advance tractable inference, it's been a blast!. with @priyankjaini @kerstingAIML @wellingmax @ahc_correia @ropeharz @eric_nalisnick @antonuccialex @guyvdb @Sriraam_UTD @andrewgwils Xiaoting and Misha
2
6
79
The poster fest I organised was so lively and saw many students showing up to rehearse for the upcoming @NeurIPSConf . Thanks #GAIL the #Generative #AI Lab @InfAtEd for sponsoring the event!
1
8
84
Consider submitting works on normalizing flows, probabilistic circuits and tractable inference at large at the 4th Tractable Probabilistic Modeling Workshop #TPM2021 to be held at @UncertaintyInAI #UAI2021!. Deadline: 28 May. Please RT and share! 👇👇👇.
2
32
78
Congrats to Dr. @andreasgrv who just defended his #PhD thesis on unargmaxability and constraints in deep learning, examined by Vlad Niculae and @driainmurray 🎉. A pity you could not (yet) see Andreas' beautifully designed thesis!. In the meantime, some pics from the pre-viva 👇
6
10
79
Mixtures are *A M A Z I N G*!. They let you:.1) marginalize in O(k) (k = # components) if components can marginalize (eg. Gaussian).2) approximate well enough *any* density for k → +∞. i.e., asymptotically. but in practice? . How many Gaussians to fit the densities in the pic?
5
11
76
when they ask me why I won't be going back to #Italy anytime soon. a Vatican priest with a PhD in moral theology and some interests in ethical aspects of AI is now leading the Italian task force of #AI (previously led by a 90 years old guy with no experience in computer science).
Il teologo francescano Paolo #Benanti è il nuovo Presidente della Commissione sull'Intelligenza Artificiale per l'informazione, dopo le dimissioni di Giuliano #Amato. @ultimora_pol
6
11
72
controversial opinion: this year's #NeurIPS2021 reviews are the best I ever had. (how is this possible?).
15
2
71
I am Vienna for @iclr_conf #ICLR2024 for a week presenting work done with @loreloc_ @diegocalanzone & Christian Jimenez. Ping me if you want to chat about . - realiable and complex probabilistic reasoning 🎯📊.- neuro-symbolic integration 🤖🧠.- fast inference⚡️🔬.
Mixture models that subtract probability density can be exponentially more expressive in representing complex distributions. We prove this when they are learned by squaring tensorized networks, and show how they represent a unifying modelling framework in our #ICLR2024 spotlight!
2
12
69
thanks @laurent_dinh for sharing more than just research successes and achievements at @iclr_conf #ICLR2020: a more human and fuller journey from rejections at conferences to collective advances as a community
0
5
66
join @loreloc_ #ICLR2024 @iclr_conf at Poster #210 .Thursday 10.45 (today in 20 mins!) . to learn how to "cut" distributions while preserving distribution semantics and tractability
Mixture models that subtract probability density can be exponentially more expressive in representing complex distributions. We prove this when they are learned by squaring tensorized networks, and show how they represent a unifying modelling framework in our #ICLR2024 spotlight!
0
9
67
ready to talk on how you can integrate #symbolic and #logical #constraints in #deep #learning #ML via #circuits, flawlessly train end2end and generalize many #nesy works at @Uni_Stuttgart . w/ @ststaab @BoXiongs @Mniepert and @ELLISforEurope . slides:
1
11
63
De-hyping #ML and #AI, and super-human performance claims + (finally!) treating ImageNet as multi-label classification. I am in love with the rigorous and skeptikal empirical science by @Vaishaal Rebecca Roelofs @HoriaMania @beenwrekt @lschmidt3. 👇👇👇.
3
15
61
@CrazyItalianPol Winx are a national treasure, the second most successful product we exported, after fascism.
1
4
58
Superhappy to discuss the Atlas paper. 📜 tomorrow at the @MITCSAIL #Programming #Languages Review. 👉 as one of the papers that "that may substantially transform the #PL community and beyond"!
1
9
63
Preparing slides for tomorrow's talk and this popped up
0
3
57
Miracolo!. Just 13 hrs before the #NeurIPS20 deadline Our Lady of the Well in Capurso (La Madonna del Pozzo di Capurso) appeared to show me what were wrong in the experiments!. Now we have a submission we can submit to @NeurIPSConf we just need to rerun all and write a paper!.
7
1
57
@shaohua0116 honestly. this tells a long story about #authorship and the current state of #ML and #AI fields. 32 papers is 1 paper every 12 days.20 papers is 1 paper every 18 days.16 papers is 1 paper every 23 days.
1
4
54
if this book is half as good as @jmtomczak series of blog posts, it is a must buy for everyone doing a PhD/research in #unsupervised #ML and #deep #generative models!.
I am excited to announce that my book, "Deep Generative Modeling", is available online and in print (@SpringerNature): Code used in the book is freely available online: (1/4)
1
5
53
still 💯s of @NeurIPSConf mugs left!.what's wrong with you people?. (you can grab one more for your parents at the reception)
12
2
57
VAEs, RAEs and more in 🐍!.
After 8 months of long coding nights ☕️ we finally officially release Pythae 🥳, a python library unifying generative autoencoder implementations including vaegan🥗, vqvae or RAEs. 🖥️ github repo: 👉paper:
0
6
51
A good metaphor for some @iclr_conf #iclr2021 submissions I came across 👇👇👇. Some good ingredients (ideas) are strangely placed on a plate (literature) and hidden by an overabundance of sauce (details, claims) no one is going to eat through (hard to check). Bon appetit!
6
3
50
"I cannot emphasize enough that you should not interpret #Bayesian methods as just regularizers of maximum likelihood #MLE learning". @andrewgwils in the "Bayesian Deep Learning and a Probabilistic Perspective of Model Construction" at @icmlconf . 👇👇👇.
1
7
55
Thank you for your response. I will keep my score.
6
0
52
""".There is a tendency to classify work as Bayesian or not Bayesian, with very stringent criteria for what qualifies as Bayesian [. ] We believe this mentality encourages tribalism, which is not conductive to the best research.""".words of wisdom by @andrewgwils @Pavel_Izmailov.
Our new paper "Bayesian Deep Learning and a Probabilistic Perspective of Generalization": Includes (1) benefits of BMA; (2) BMA <-> Deep Ensembles; (3) new methods; (4) BNN priors; (5) generalization in DL; (6) tempering in BDL. With @Pavel_Izmailov. 1/19
1
9
47
Germany 🇩🇪, Italy 🇮🇹 and Austria 🇦🇹, wtf?. do you really need to see Kyiv carpet bombed before deciding to effectively isolate Putin and exclude Russia from SWIFT?.
2
2
49
A couple more lectures for the #MLPR course on probabilistic #ML @InfAtEd next week. While going through the recordings of past lectures, I found your cube materializing on my slides, @ZoubinGhahrama1!
2
5
54
On this day, 422 years ago, Giordano Bruno was burned alive by the Catholic Church for saying that stars were distant suns surrounded by planets, and that Earth was not the center of this infinite universe. #GiordanoBruno #1600.
4
7
51
5 postdoc positions in #AI and #ML in #AI4Science, Climate and Earth Sciences at the Image and Signal Processing group @isp_uv_es in Valencia, Spain 🇪🇸🇪🇺. 👇👇.
1
24
53
my advice to #PhD applicants👇 . write a #research statement to highlight how your past experience & current interests in #ML overlap with those of the labs you are applying to!. I'll be hiring via #ELLIS too, you can read about research @ #APRIL here 👉
Just a few days are left until our #ELLISPhD application portal will close for this year! Take a look at our FAQs if you still have questions and don’t forget to submit your application by November 15! #machinelearning #AI #phdprogram @ai_elise
0
11
51
The #Edinburgh @InfAtEd crew at @NeurIPSConf decompressing!. including .@henrygouk @loreloc_ @EmilevanKrieken.@LaurenNDeLong @bryanlimy.@vsimkus. and with bonus photo bombing by @jmtomczak!
2
4
53
we evaluate #continual learning models on "baby" benchmarks, where it is easy to show no catastrophic forgetting!. we propose a new simple benchmark that glues simple but still challenging tasks in a curriculum: from MNIST to Imagenet and back!. 📜
1
15
54
We are about to start the 5th Workshop on Tractable Probabilistic Modeling #TPM2022 at @UncertaintyInAI 🚀🚀🚀. @martin_trapp @yjchoi012 @fabri_ven @eric_nalisnick
2
2
52
A gentle introduction to #Probabilistic #Circuits #PCs and the importance and joys of #tractable #inference and #modeling in #ML and #AI . A new tutorial with @yjchoi012 @ropeharz @guyvdb at @ECAI2020 #ECAI2020 starting in ~30 mins👇👇👇.
1
10
51
One week to the 5th #TPM workshop!.Pass by if you are attending @UncertaintyInAI or in #Eindhoven. We have a stellar list of accepted papers. 👉 and best paper awards! 🏆🏆🏆. Thanks to our sponsor @GoogleAI @ZoubinGhahrama1
1
12
51
15 authors and no one who knows that pretty tables have no column separators. (ps. is this a new trend to have only abstract and figures on the 1st page?). (pps. did we really need to coin the term "Foundation Transformers"?).
4
2
52
Twitter people, does anyone have the full email address of Prof. Karl Marx? 👇👇👇. I need to ask him something urgently.
4
3
46
#Nordic @probabilisticai Summer School kicks in, what an amazing crowd of #PhD students who want to learn more about #Probabilistic #ML #AI. with Silja Renooij about to give part II of the intro to probabilistic modeling
0
7
50
factorizing a tensor with infinite dimensions⁉️.yes that's possible‼️. and this gives you tractable inference, and the simplest and most principled way to train probabilistic circuits at scale✅. spotlight at #NeurIPS24. 📜 .🖥️
I’ll be attending #NeurIPS2024, where I’ll present our spotlight: Scaling Continuous Latent Variable Models as Probabilistic Integral Circuits (PICs). TL;DR: We learn continuous hierarchical mixtures as DAG-shaped PICs, and scale them using neural functional sharing techniques.
0
6
53
About to fly to #Germany for a short #research tour of @SIC_Saar @ELLISforEurope @Uni_Stuttgart @CS_TUDarmstadt . Super happy to meet old friends and colleagues @IValeraM @ststaab @Mniepert @bschoelkopf @PhilippHennig5 @MatthiasBethge @kerstingAIML @javaloyML and more!
5
1
49
#ICML2023 @icmlconf deadlines are out:. 🚨Paper Submissions Open on @openreviewnet:.🗓️Jan 09 '23 02:00 PM UTC. 🚨Full Paper Submission Deadline:.Jan 26 '23 08:00 PM UTC.
0
7
50
out of local minima, it is important to encourage risk and recognize that new approaches can't initially yield state-of-the-art competitive results. Nor are they always sold according to the recipes we are used to. [2/2].""". thanks @NeurIPSConf for the reminder 🫶.
2
2
50
AISTATS is one of those venues in #AI and #ML with high quality in reviews, submitted content and discussions!. Plus, #AISTATS2022 will be live in Valencia, Spain!.
We invite submissions to the AISTATS 2022 conference and welcome paper submissions on artificial intelligence, machine learning, statistics, and related areas!.
1
10
50
Submit your work on normalizing flows, probabilistic circuits and other tractable models to #TPM2022 the 5th Workshop on #TPMs to be held at #UAI2022 @UncertaintyInAI!. Deadline: June 9. with @yjchoi012 Fabrizio Ventola @martin_trapp @eric_nalisnick. 👉
0
16
48
We are organizing a Dagstuhl Seminar (@dagstuhl) in April on . "Recent advancements in #tractable probabilistic inference". to exactly discuss where we stand and what the future looks like for #deep generative models!. with @priyankjaini @kerstingAIML @wellingmax . Stay tuned.
Are there any flexible probabilistic model classes (neural or non-neural) that permit tractable computation of expectations, information & distributional properties generally (bonus points for conditional variants) @guyvdb @martin_trapp @DaniloJRezende ?.
0
12
44
@thegautamkamath it is to counterbalance all the fake Theorems we put in NeurIPS papers every year.
1
1
45
🚨Is complex query answering really complex?🚨. unfortunately not!. the current benchmarks boil down to link prediction 98% of the time. how to fix this???. 👇👇👇.📜 with @c_gregucci @BoXiongs @loreloc_ @PMinervini @ststaab
@michael_galkin yes! . the way missing edges are placed is the key ingredient to complexity in CQA. 🚨🚨🚨.Unfortunately, all current benchmarks are baby benchmarks requiring to predict a single missing edge for 99% queries. Bottom line: we are just evaluating link prediction.
2
15
56
Even more remarkable than the publicly available lib in 🐍 are the benchmarks run in the paper (👉 . TL;DR: deterministic AEs can be comparable and often much better generative models than VAEs and variants!. A short 🧵👇.
VAEs, RAEs and more in 🐍!.
4
10
45
6
0
45
0
0
45
I am thrilled to announce that 2/2 bars have been accepted by this COVID test, making it one of the most positive results it can provide. (no worries, I will be back to boast about #NeurIPS hopefully soon)
9
0
46
congrats to Dr. @EmilevanKrieken who successfully defended his #PhD thesis . "#Optimization in #Neurosymbolic #Learning Systems" .( . obtaining a cum laude from @VUamsterdam!. #ML #AI #NeSy.
8
1
42
if this isn't neuro-symbolic, then what is?. (do we need "test-time compute" as a new name to the zoo?)
Just 10 days after o1's public debut, we’re thrilled to unveil the open-source version of the groundbreaking technique behind its success: scaling test-time compute 🧠💡 . By giving models more "time to think," LLaMA 1B outperforms LLaMA 8B in math—beating a model 8x its size.
2
4
43
This paper will get a thread at a certain point as it has many stories to tell, beyond its interesting technical content. For example, it was brutally rejected at @icmlconf earlier this year before getting accepted as oral now at @NeurIPSConf #NeurIPS2021! 👀.
💥Oral💥. "A Compositional Atlas of Tractable Circuit Operations for Probabilistic Inference". by A Vergari @tetraduzione YooJung Choi @yjchoi012 Anji Liu @liu_anji Stefano Teso @looselycorrect Guy Van den Broeck @guyvdb. 📜:💻: 8/8
1
2
43
teaching and doing research the same semester be like.
2
1
43
Amazing art and explanation of the underlying math. A mixture of increasingly more complex polynomials to "fit" a realistic terrain!🔥. This can be a nice recipe for a boosting algorithm for density estimation. 👀.
0
5
42
don't miss this if you. - are in #Vienna for #ICML2024 .- are interested in #neurosymbolic #AI.- still believe that "logic is non-differentiable!".
I'll be giving a talk at the ICML 2024 Differentiable Almost Everything Workshop!. ⏲️ Friday 9:10 in Stolz 0. Join us if you're interested in learning about differentiating through logic and other algorithms!.
1
6
43
what is going on with @NeurIPSConf early registration for conference and workshop costing 850$?. Isn't it almost double the pre-pandemic cost of an in-person conference?
4
6
40
