emaballarin
PhD student in (neuro-inspired) Deep Learning. Also: robust / trustworthy / Bayesian DL, kernel methods, computational neuroscience. Overenthusiastic tinkerer.
Dept. of Maths, UniTS | @LACoNIC-UniTS ⊆ @ailab-units | @sissa-data-science | @AI-Student-SocietyTrieste, Italy
Pinned Repositories
CARSO
👀🛡️ Code for the paper “Carefully Blending Adversarial Training and Purification Improves Adversarial Robustness” by Emanuele Ballarin, Alessio Ansuini and Luca Bortolussi (2024)
deeplearning-units
🧠🧑🏻💻 Repository for the Deep Learning course labs/practica (UniTS, Spring 2024)
ebtorch
🤓🔥 Collection of PyTorch additions, extensions, utilities, uses and abuses
fastwonn
🏁🥈Fast, GPU-friendly, differentiable computation of Intrinsic Dimension via Maximum Likelihood, the TwoNN algorithm, and everything in between!
foveatorch
👀🔥 Differentiable foveated vision for Deep Learning methods
HinTorch
A critical, 🔥PyTorch-flavoured attempt at reproducing the results of 📃"Learning representations by back-propagating errors" (Nature, 1986) by Rumelhart, Hinton and Williams.
nmc-numpyro
🎲🔥 Unofficial implementation of Newtonian Monte Carlo MCMC sampling algorithm (after Arora et al., 2020) in JAX/NumPyro, with full NumPyro-API compatibility
persaturam
Because using gists is good... until they become too many to manage!
probml-units
🎲🧑🏻🏫 Repository for the Probabilistic Machine Learning labs/practica (@ UniTS, Spring 2023)
wsl2-linux-kernel-clearsky
[AUTO-RELEASE] Sources of latest WSL2 Linux kernel from Microsoft, with graysky2's patchset
emaballarin's Repositories
emaballarin/ebtorch
🤓🔥 Collection of PyTorch additions, extensions, utilities, uses and abuses
emaballarin/CARSO
👀🛡️ Code for the paper “Carefully Blending Adversarial Training and Purification Improves Adversarial Robustness” by Emanuele Ballarin, Alessio Ansuini and Luca Bortolussi (2024)
emaballarin/foveatorch
👀🔥 Differentiable foveated vision for Deep Learning methods
emaballarin/carso-msc-thesis
LaTeX source and related material for my M.Sc. thesis in Deep Learning
emaballarin/adv-ml-units
Repository for the Advanced Topics in Machine Learning practica (@ UniTS, Fall 2024)
emaballarin/homebrew-cloudflare
[🛠️🧪]
emaballarin/nupic.torch
[🛠️ Tweaks] Numenta Platform for Intelligent Computing PyTorch libraries
emaballarin/paper2remarkable
[🧪] Fetch an academic paper or web article and send it to the reMarkable tablet with a single command
emaballarin/robustbench
RobustBench: a standardized adversarial robustness benchmark [NeurIPS'21 Benchmarks and Datasets Track]
emaballarin/TabPy
🧪 [Case Sensitive Config Parsing] Execute Python code on the fly and display results in Tableau visualizations:
emaballarin/xberkmono
Berkeley Mono NerdFonts Complete patching machinery
emaballarin/adversarial-attacks-pytorch
🩹 PyTorch implementation of adversarial attacks [torchattacks]
emaballarin/apex
[Personal fork] NVIDIA Apex
emaballarin/chrome-open-in-spotify-client
Instead of opening open.spotify.com links in the browser they will always be opened in Spotify
emaballarin/confidenceinterval
🩹 The long missing library for python confidence intervals
emaballarin/DADApy
[⏩] Distance-based Analysis of DAta-manifolds in python (for use within the pipfury.ballarin.cc PyPI index)
emaballarin/digitalgarden_template
[⚒️ Clone of the original, with updated dependencies]
emaballarin/einshape
🩹
emaballarin/gemini-ailab-units
A Beamerposter template with AILab @ UniTS logo and colors. Forked from Gemini.
emaballarin/hdbscan
A high performance implementation of HDBSCAN clustering.
emaballarin/iris
[🍽️] FORK: Iris is a simple utility written completely in python which allows to sync folders between local and remote ssh (sftp) server
emaballarin/jmp
🩹 JMP is a Mixed Precision library for JAX.
emaballarin/jupyter-notebook-viewer
[🧪] Google Chrome extension for viewing Jupyter Notebooks in the browser, without a running Jupyter Server
emaballarin/mkcert
🍽️ A simple zero-config tool to make locally trusted development certificates with any names you'd like.
emaballarin/ml_dtypes
[🩹 UN-DEPRECATE] A stand-alone implementation of several NumPy dtype extensions used in machine learning.
emaballarin/ReVancedBuilder
emaballarin/richer_tqdm
⚡Improved tqdm.rich decorator for iterators and associated machinery
emaballarin/schedule_free
[🔀] Schedule-Free Optimization in PyTorch
emaballarin/torch_jax_interop
🔀 Simple tools to mix and match PyTorch and Jax - Get the best of both worlds!
emaballarin/wsl-hello-sudo
PAM-based sudo authentication by Windows Hello on Windows Subsystem for Linux (WSL)