Mohamed Elsayed

Dept. of Computing Science, University of Alberta

I am a PhD student at the RLAI lab working with Rupam Mahmood and affiliated with Amii. I develop general mechanisms for continual learning, focusing on enabling deep reinforcement learning methods to learn from continuous streams of experience, particularly for resource-constrained embodied agents.

Prior to this, I completed my master's degree from the University of Alberta and I completed my bachelor's degree in Communication and Information Engineering from Zewail City of Science and Technology, Egypt.

Taking photos is something I enjoy doing when I have some free time. You can check out the world as I see it from here. I also love observing how animals behave and interact with their environments, especially those intelligent ravens.

news

Oct 18, 2024 I am excited to share that our paper on streaming deep reinforcement learning is now released! You can check the code from here.
Sep 25, 2024 Our paper led by Gautham Vasan on deep policy gradient methods without batch updates, target networks, or replay buffers is accepted at NeurIPS 2024.
May 15, 2024 Our paper on the role of weight clipping for deep continual and reinforcement learning is accepted at RLC 2024.
May 1, 2024 Our paper on revisiting scalable Hessian diagonal approximations for applications in reinforcement learning is accepted at ICML 2024.
Jan 16, 2024 Our paper on loss of plasticity and catastrophic forgetting is accepted at ICLR 2024.
Sep 1, 2022 Started my PhD in continual representation learning.
May 13, 2022 Defended my MSc thesis on online representation search.
Dec 31, 2020 Finished my internship at Huawei, Edmonton (see this NeurIPS workshop paper for a summary)
May 1, 2020 Started my internship at Huawei, Edmonton.