Sébastien Arnold
Los Angeles, California, United States
Summary
🤩
Rockstar🎓
Top SchoolSébastien Arnold is a machine learning researcher and engineer with 13 years of experience building meta-learning and deep learning systems across industry labs and academia. He has interned on cutting-edge projects at Google, AWS, and Amazon focused on meta-learning for large language models, episodic sampling, and causal representation learning, and contributed to open-source ML tooling such as integrating MAML into the popular learn2learn PyTorch meta-learning library and improving data loaders in Intel’s neon framework. With a PhD in Machine Learning from USC and research stints at Mila and ETH Zürich, he blends rigorous theory (variance reduction, online stochastic optimization) with practical implementation and reproducible examples that bridge toy problems and RL environments. Based in Los Angeles, he prefers contact via his website and brings a pattern of shipping research code that scales from benchmark datasets to simulated control tasks—an indicator of both experimental depth and engineering polish.
13 years of coding experience
2 years of employment as a software developer
Bachelor's degree, Mathematics and Computer Science, Bachelor's degree, Mathematics and Computer Science at University of Southern CaliforniaMaturité Fédérale, Maturité Fédérale at Gymnase Auguste PiccardBachelor of Science - BS, Computer Science (Transferred to USC), Bachelor of Science - BS, Computer Science (Transferred to USC) at ETH ZürichHigh-School Summer Session, High-School Summer Session at Stanford University
French, English, Italian, German
Github Skills (13)
data-loading10
pytorch10
machine-learning10
deep-learning10
python10
load-data10
few-shot-learning10
numpy10
data-set10
datasets10
testing9
reinforcement-learning9
documentation7
Programming languages (10)
Github contributions (5)
learnables/learn2learn
Aug 2019 - Aug 2022
A PyTorch Library for Meta-learning Research
Role in this project:
ML Engineer
ML EngineerContributions:13 releases, 9 reviews, 388 commits in 3 years 1 month
Contributions summary:Sébastien implemented and integrated the MAML (Model-Agnostic Meta-Learning) algorithm into the PyTorch meta-learning framework, Learn2Learn. They added the MAML algorithm, demonstrated its usage in a toy example using the MAML algorithm with a Normal distribution, and further extended the capabilities by integrating the MAML algorithm with RL environments such as Ant and Half-Cheetah. Moreover, they added a new task generator API and converted the existing examples to it.
pytorchmeta-learningmetalearningmeta-optimizationdeep-learning
NervanaSystems/neon
May 2015 - Jul 2015
Intel® Nervana™ reference deep learning framework committed to best performance on all hardware
Role in this project:ML Engineer
ML EngineerContributions:10 commits, 7 PRs, 19 comments in 1 month
Contributions summary:Sébastien primarily contributed to the development and enhancement of the CIFAR100 dataset loader within the Intel Nervana deep learning framework. They added the CIFAR100 loader, updated the documentation, and refactored the code to move coarse labels to kwargs. These modifications involved data loading and preparation for machine learning tasks, directly improving the framework's capabilities. The user also fixed issues related to MNIST and CIFAR100 testing.
deep-learningbest-performanceintelmachine-learningperformance
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.