Alfredo Canziani
Assistant Professor Of Computer Science at Courant Institute of Mathematical Sciences
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolAlfredo Canziani is an assistant professor of computer science at NYU’s Courant Institute with 13 years of experience bridging deep learning research, engineering, and teaching. He focuses on ML for autonomous driving — from uncertainty estimation and failure detection to latent forward models for long-term planning — while contributing practical implementations to open-source projects like Torch autograd, Torch/PyTorch demos, and NYU deep learning course notebooks (autoencoders/VAEs, regularization experiments). He co-taught Yann LeCun’s deep learning class and is known for turning complex theory into interactive, multimedia pedagogy. Trained with a PhD from Purdue and top honors from Università di Trieste, he pairs rigorous math with creative interests as a musician, dancer, and cook, giving his research a distinctive interdisciplinary and human-centered flavor.
13 years of coding experience
7 years of employment as a software developer
Purdue University
Italian, English, Spanish, Chinese, American Sign Language, Slovenian
Stackoverflow
693reputation
63kreached
3answers
17questions
Github Skills (32)
pytorch10
python10
image-processing10
machine-learning10
recurrent-neural-networks10
autoencoders10
autoencoder10
lua10
keras10
recurrent-neural-network10
deep-learning10
neural-networks10
computer-vision10
neuralnetwork10
jupyter-notebook10
Programming languages (29)
Github contributions (5)
Atcold/NYU-DLSP20
Jul 2018 - Apr 2022
NYU Deep Learning Spring 2020
Role in this project:
Data Scientist
Data ScientistContributions:1 release, 405 reviews, 399 commits in 3 years 9 months
Contributions summary:Alfredo added a Keras-based notebook for regularisation in neural networks, exploring different regularisation techniques such as L2 regularization, L1 regularization, and dropout. They also included data loading and preprocessing steps utilizing the IMDB dataset and tokenization. Furthermore, the user implemented a model to study regularisation and visualised train/test loss and accuracy, suggesting a focus on understanding and demonstrating the impact of regularisation on model performance.
pytorchdeep-learningjupyter-notebookneural-netsmachine-learning
Atcold/NYU-DLSP21
Apr 2021 - Aug 2022
NYU Deep Learning Spring 2021
Role in this project:ML Engineer
ML EngineerContributions:133 reviews, 37 commits, 37 PRs in 1 year 4 months
Contributions summary:Alfredo primarily contributed to the development and experimentation of deep learning models within the repository. Their work included the addition of an autoencoder and VAE implementations, leveraging PyTorch. The user also made bug fixes and updated existing notebook files.
ebmdeep-learningspringspring-learningnyu
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.