Name: Esra'a
Type: User
Company: MILA / University of Montreal
Bio: Computer Scientist with an obsession for new challenges :)
Researcher in AI/ML/RL @mila-iqia / UdeM. formerly, @AmiiThinks @rlai-lab, UAlberta. Loves skiing.
Location: Canada
Blog: esraasaleh.com
Esra'a's Projects
Implementation of https://srush.github.io/annotated-s4
Essentially a personal sandbox. A compilation of classical rl agents for control that are integrated with the rl-control-template
CLEVR-Robot: a reinforcement learning environment combining vision, language and control.
CUHacking 2018 asking app
The exoplanet explorer is a website project that aims to provide a 3D visualization experience along with filtering of real data of all exoplanets discovered by the Kepler mission.
A compilation of data-informed articles on the novel coronavirus, focusing on impacts/strategies for public health, with careful consideration to source credibility.
My compatibility fixed deepkoopman
A Python tool to export annotations and highlights from a Kobo SQLite file.
Web app to communicate with instructors in classroom
Magenta: Music and Art Generation with Machine Intelligence
Jax based implementations of rl algorithms with meta-learning
MiniGrid Implementation of BEHAVIOR Tasks
A set of HTML widgets that could be embedded into Notion.so https://www.notion.so/ pages. For more see https://blog.shorouk.dev/notion-widgets-gallery/
Optax is a gradient processing and optimization library for JAX.
This is code for our project that investigated the effect of reward sparsity on the differentially private reinforcement learning algorithm by Wang and Hegde (2019).
[NeurIPS'21 Outstanding Paper] Library for reliable evaluation on RL and ML benchmarks, even with only a handful of seeds.
An AI library written in Rust
Physics games involving tool use for studying human physical representations and rapid trial-and-error learning
A thesis template for master's/doctoral thesis.
Project at Terrible Hacks for waking someone up with an innovative alarm