
Postdoc @PontilGroup @IITalk. Ex: @UniTrieste, @ai_ucl. Representations, robustness, empirical methods, kernels, physics & neuroscience. Quantitativist.
Genova (Italy)
Joined March 2014
- Tweets120
- Following5.5K
- Followers934
- Likes2.4K
Iscritto al Registro delle Opposizioni, vengo contattato in modo non richiesto da @TIM_Official per pubblicità. Lo faccio notare all'operatore e ricordo la sanzione irrogata dal Garante per la Privacy. La risposta: "Vabbè, prenderemo un'altra multa!". È la vs posizione ufficiale?
Emanuele Ballarin retweeted
After the holiday break, we are ready to kick-off our activities for the new year! Stay tuned...
Thanks to everyone who supported us until now and helped us grow; and looking forward to meet more people and cross more paths in the times to come.
To everybody, a happy 2022!
Unmissable!
An actual superstar in the #DeepLearning field, and a world-class educator, @alfcnz needs no further introduction. We are glad and honored to host him for our #XmasSeminar about Energy-Based Self-Supervised Deep Learning.
For those not in Trieste: youtu.be/bDvpuaPq8Vc
Emanuele Ballarin retweeted
Are you looking for MSc thesis / research internship / PhD opportunities in the fields of #datascience #ai #ml #computationalscience?
Speak directly to potential supervisors from @UniTrieste, @Sissaschool, ICTP, INAF during our AI2S Research Fair 2022!
ai2s.it/events/ai2s-research…
Emanuele Ballarin retweeted
Pyro 1.0 is released: stable APIs, a jit-compatible Predictive helper, new normalizing flows, parallel-scan GPs and state space models, more-automatic AutoGuides, an OED tutorial, FoldedDistribution, and PyroModule[-] to make your nn.Modules Bayesian github.com/pyro-ppl/pyro/rel…
Emanuele Ballarin retweeted
It's official! @PyTorchLightnin is part of the @PyTorch ecosystem now - Keras-like ML library for researchers.
Check it out if you haven't yet!
github.com/williamFalcon/pyt…
@MILAMontreal @AmiiThinks @berkeley_ai @stanfordnlp @MIT_CSAIL @DataScienceCtrl @DeepLearningHub @hmason
Emanuele Ballarin retweeted
Level 4 Autonomous Driving: Mostly automated, occasional intervention
Level 5: Fully autonomous, no intervention
Level 6:
This year, a Christmas present in advance!
Replying to @glouppe
This year, I am really satisfied with the second version of my Introduction to AI lectures. Tomorrow will already be the last lecture! All (700+!) slides can be downloaded at glouppe.github.io/info8006-i…
This is just great!
My deep-learning course materials are now updated to PyTorch 1.0.0.
fleuret.org/ee559
Close to 900 slides! Comments are welcome.
#deeplearning #idiap #epfl #pytorch
Emanuele Ballarin retweeted
The gradual take over of physics... we got neutrinos into the NeurIPS logo. Is that a right-handed neutrino? – at New Jersey, USA
Hey guys / ladies, here you have the new logo. Feel free to use them in your presentations. 😇 #NeurIPS2018 drive.google.com/drive/folde…
Verifying myself: I am emaballarin on Keybase.io. ftWeEHUubUtrPytJYj-AZvAHw4Ab7Q0uBiqS / keybase.io/emaballarin/sigs/…
A sneak peek through a neural classifier's "eyes"...
Modern #MachineLearning just looks like magic, if you carefully define "magic" as stacked, nonlinear-ized affine transformations.
A must-read thread... and probably the most interesting thing you'll read today.
(for reference, at the moment it's 22 tweets long)
1/ Who wants to hear some scientific intrigue?
A few weeks ago, a group of physical chemists posted a paper online announcing the observation of superconductivity at room temperature.
Today I posted a comment pointing out something funny in their data.
arxiv.org/abs/1808.02929
Emanuele Ballarin retweeted
Standardizing a Machine Learning Framework for Applied Research bit.ly/2Ktq69N #AI #DeepLearning #MachineLearning #DataScience
To-do list for you reading:
1) Allocate ~7 hours in the future;
2) Follow this #PyTorch & #DeepLearning crash-course.
This is what's called "time well-spent"!
It's been an honour teaching deep learning this week at @Princeton! All lecturing material is available here: github.com/Atcold/PyTorch-De… princeton.edu/news/2018/07/2…
Emanuele Ballarin retweeted
Ablation studies are crucial for deep learning research -- can't stress this enough.
Understanding causality in your system is the most straightforward way to generate reliable knowledge (the goal of any research). And ablation is a very low-effort way to look into causality.
Emanuele Ballarin retweeted
Because of #GDPR, USA Today decided to run a separate version of their website for EU users, which has all the tracking scripts and ads removed. The site seemed very fast, so I did a performance audit. How fast the internet could be without all the junk! 🙄
5.2MB → 500KB


























