Maryam Hashemzadeh
Research Assistant at Mila - Quebec Artificial Intelligence Institute
Montreal, Quebec, Canada
Summary
👤
Senior🎓
Top SchoolMaryam Hashemzadeh is a research-focused machine learning scientist with eight years of experience bridging large language models, reinforcement learning, and lifelong learning from Montreal’s Mila. She currently advances Mixture of Experts architectures and investigates efficient knowledge integration and removal, combining RL techniques like GRPO to reduce hallucinations and improve LLM reliability. Her work aims to align model reasoning and long-term adaptability with human-like thinking, emphasizing interpretability and trustworthiness as much as performance. With research stints at Amii, University of Alberta (RLAI), ATR in Japan and visiting roles at ServiceNow Research, she brings a global, interdisciplinary perspective to applied ML problems. Maryam’s background in both academic and applied research contexts means she routinely translates theoretical ideas into experiments and tools that move models toward continual, robust learning.
9 years of coding experience
5 years of employment as a software developer
Master of Science - MS, Computer Science, Master of Science - MS, Computer Science at University of AlbertaDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Université de Montréal
Github Skills (16)
sgd6
tensorflow5
lstm5
pytorch5
language-model4
reinforcement-learning4
standardized3
text-based3
transformer3
llm2
science2
machine-learning2
virtual-environment2
peft1
react1
Programming languages (3)
Github contributions (5)
chandar-lab/SubGoal_Distillation_LLM
May 2024 - Jul 2024
Code for paper Sub-goal Distillation: A Method to Improve Small Language Agents, accepted at CoLLAs 2024.
Contributions:13 reviews, 2 PRs, 9 pushes in 2 months
decision-transformersllmreinforcement-learningtransformer
MHashemzadeh/Offline-Online-RL
Nov 2021 - Jul 2022
Contributions:13 pushes, 9 branches in 8 months
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.