DPhil in Engineering Science, University of Oxford, 2018-2022
M.Eng in Engineering Science, University of Oxford, 2014-2018
2021 - Present: Senior Researcher - Microsoft
I work on compressing neural networks for quick computation on edge devices - specifically I have been working on compressing Language Models, and Diffusion based networks for both language and image generation
July - December 2021: Research Intern - Microsoft
I worked on compressing especifically superresolution models for cheap computation on edge devices - this included using a pipeline consisting of Distillation, Pruning, and Quantization
July 2019: Graduate Engineer - Omnitek (now sold to Intel)
I worked on Quantizing Neural Networks for efficient computation on FPGAs
Python, PyTorch, C++
Ashkboos, S., Croci, M. L., Nascimento, M. G. D., Hoefler, T., & Hensman, J. (2024). SliceGPT: Compress Large Language Models by Deleting Rows and Columns. arXiv preprint arXiv:2401.15024.
Gennari do Nascimento, M., et al. HyperBlock Floating Point: Generalised Quantization Scheme for Gradient and Inference Computation. IEEE, 2023, pp. 6353–62.
Gennari do Nascimento, Marcelo, Theo W. Costain, and Victor Adrian Prisacariu. "Finding non-uniform quantization schemes using multi-task gaussian processes." European Conference on Computer Vision. Cham: Springer International Publishing, 2020.
Nascimento, Marcelo Gennari do, Roger Fawcett, and Victor Adrian Prisacariu. "Dsconv: Efficient convolution operator." Proceedings of the IEEE/CVF International Conference on Computer Vision. 2019.