Skip to content
View pbonazzi's full-sized avatar
📈
📈
10 followers · 6 following

Achievements

Achievement: YOLOAchievement: Pull Sharkx2Achievement: Quickdraw
BetaSend feedback

Achievements

Achievement: YOLOAchievement: Pull Sharkx2Achievement: Quickdraw
BetaSend feedback
Block or Report

Block or report pbonazzi

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
pbonazzi/README.md

👋🏻 I am a Machine Learning Research Scientist, Engineer at ETH Zurich. With a strong background on Statistics and half a decade of experience in training models.

🔥 Currently, I'm working on Efficient Machine Learning (QAT, Spiking Neuron Networks, Hardware-Aware Network Design). To this end, I interface a lot with hardware especially with Neuromorphic Chips (Sinabs, Lava, etc.) , GPUs, ASICs (IMX500, Coral, etc.) and FPGAs.

⏳ Previously, I worked on diffusion models, differentiable rendering, vision-language model to reconstruct and de-noise 3D point clouds from images. I also worked on conditional 3D scene generation architectures from scene graphs and designed my own graph transformer from scratch (back in 2021).

❤️ Passions : AI, walking, public speaking.

Pinned

  1. retina retina Public

    [CVPRW 2024] Retina is an eye tracking method suitable for deployment on a neuromorphic system on chip.

    Python 15 1

  2. tinytracker tinytracker Public

    [SENSOR 2023] TinyTracker is a highly efficient, fully quantized model for 2D gaze estimation designed to maximize the performance of the edge vision systems .

    Python 7

  3. guassian-diffusion guassian-diffusion Public

    Few-shot point cloud reconstruction and denoising via learned Guassian splats renderings and fine-tuned diffusion features.

    Python 1

  4. graph-scene-transformer graph-scene-transformer Public

    3D Scene Generation from Scene Graphs and Graph Self-Attention

    Jupyter Notebook 1

  5. network-cascade-failure network-cascade-failure Public

    Analysis of interdependent networks (Erdos–Renyi , Scale-free and Paris Multilayer Transportation Network).

    Jupyter Notebook 13 2

  6. from-datascience-to-ai from-datascience-to-ai Public

    From Data Science to Neural Networks

    Jupyter Notebook 5