Nicola Bernini

My personal page

https://nicolabernini.github.io/

Work Experience

Currently Sr. Perception Engineer @ Zipline

Previously @ Uber Elevate and Ambarella working on Computer Vision and ML for Autonomous Vehicles, Embedded Deep Learning

Some of the things I have been working on

• at the intersection of Machine and Computer Vision on topics like SLAM, Visual Odometry, Object Detection, … for Autonomous Vehicles (cars and flying drones)

• on Reinforcement Learning for Controller Synthesis and Properties Falsification

• at the intersection of Programming Languages and Safety, exploring approaches like Transpiling

• Deep Learning on Embedded Platforms, with model compression, pruning and quantization, dealing with the SoCs architecture and on chip hardware accelerators

• also a bit with GAN for Dataset Augmentation and Sim2Real

Recent Posts

You need to turn off tracking protection

Blockchain

ML

Coding

Wandb

Moving my ML Blog on Weights and Biases in my Wandb Profile here

• ICLR2021 Papers

  • Negative Data Augmentation

GitHub

Some of my GitHub Repos

ML

• ICML 2021 - Papers Categorization, Preview and Summary

• Negative Data Augmentation (ICLR 2021) - Summary

• Awesome ML Papers Summaries

• ML Papers Analysis

  • A bunch of papers I have read and analyzed

Blockchain

TX Analysis

• Creation of an Opyn Option

• Staking in Yaxis Metavault

Contributions to DeFi

• Yaxis Metavault - Gas Efficiency

Coding

Programming Challenges

• My Solutions to Coding Challenges

• Another set of solutions to Programming Challenges

Scala

• Bubble Sort

Social

• LinkedIn

• Twitter

• Reddit

• Medium

• Blog

• My Medium Blog About Deep Learning

• Writer for Towards Data Science, some of my articles

  • Why VAE are likelihood-based generative models

  • [Paper Analysis] - Gauge Equivariant Convolutional Networks and the Icosahedral CNN

Education

• PhD in ML and Computer Vision

  • Under the supervision of Professor Alberto Broggi and Professor Massimo Bertozzi, I have worked on SLAM and Autonomous Cars at VisLab - Ambarella

• Second degree in Physics

  • Under the supervision of Professor Cristiano Viappiani, I have worked on Molecular Dynamics Simulations for Cytoglobin, regarding this paper CO Rebinding Kinetics and Molecular Dynamics Simulations Highlight Dynamic Regulation of Internal Cavities in Human Cytoglobin

Scientific Activity

Recent Pubblications

• Published at HSCC 2021: A Few Lessons Learned in Reinforcement Learning for Quadcopter Attitude Control

  • Code available on GitHub

• Submission to ICML 2021 (currently under review)

Past Pubblications

• Scholar Profile

Reviewer

• Publons

• Reviewer for IEEE Transactions on Intelligent Transportation Systems

Interests

• Machine Learning

• Computer Vision

• Embedded Deep Learning, SoC and HW Acceleration

• Blockchain, Smart Contracts

Getting in Touch

The best way to contact me: DM on LinkedIn or Twitter