Welcome! 🚀
Hi! I am Philippe Bich, a third (and final) year PhD student at Politecnico di Torino with a research focus on Low-Energy Deep Neural Networks (DNNs). After earning a Bachelor’s degree in Computer Engineering, I continued my academic journey with a Master’s degree in Mechatronics Engineering (110/110 cum laude) from Politecnico di Torino, where I developed a strong interest in Computer Vision and Robotics. During my Master’s thesis at the Boston University Robotics Lab under the supervision of Prof. John Baillieul, I decided to pursue a PhD under the guidance of Prof. Gianluca Setti. My primary research area has become the compression/pruning of neural models for computer vision, aiming to enable their deployment on resource-constrained edge and robotic platforms. I am also particularly interested in Event-based cameras, which are low-power neuromorphic sensors and offer significant advantages for real-time applications by capturing dynamic scenes with high temporal resolution and low latency. I have published papers in several prestigious conferences and journals, including ICRA and CVPR, and I am a finalist of the CVPR 2024 Event-based Eye-tracking Challenge.
News 📰
| Aug 10, 2024 | Happy to announce that my paper “Memory in Motion: Exploring Leaky Integration of Time Surfaces for Event-Based Eye-Tracking” is accepted 🔥and I will present it at the BioCAS Conference in XI’an (China) in October! 🐲 |
|---|---|
| Jul 15, 2024 | After one year we have news from IEEE TNNLS!📰 After a major review, our paper “A Multiply-And-Max/min Neuron Paradigm for Aggressively Prunable Deep Neural Networks” is now under minor review with many positive feedbacks! 🎉 You can find a preprint of this work here! |
| Jun 25, 2024 | Today, I am in Milan as a speaker at the TinyML EMEA Innovation Forum!🤌🏼 I will be presenting on how to make Vision Transformers smaller and more energy-efficient!🔋 More info here! |
| Jun 17, 2024 | Today is CVPR time, again!👁️ I am in Seattle as a finalist of the Event-based Eye Tracking Challenge! 👓 Check out our new paper that describes how the Challenge went! |
| Jun 1, 2024 | Just submitted 📨 a new version of my paper “On the Universal Approximation Properties of Deep Neural Networks using MAM Neurons” 📐 at the IEEE TPAMI Journal!🤞🏼 Preprint available here! |
| Apr 22, 2024 | Do you want to make Vision Transformers super small? Today I am in Abu Dhabi 🐫 to present my paper “Optimizing Vision Transformers: Leveraging Max and Min Operations for Efficient Pruning” at the AICAS conference! 🌴 |
| Mar 25, 2024 | First challenge for me, but we made it to the final!🏆 Our team, the Trashcoders, is leading the CVPR Challenge hosted by the “AIS: Vision, Graphics and AI for Streaming” with more than 25 teams partecipating!🥇 We are invited to the CVPR 2024 for the final stage of the Event-based Eye Tracking Challenge! 👁️👓 |
| Aug 6, 2023 | First day in Arizona, first day of MWSCAS!🏜️ I am here to present my paper “Multiply-And-Max/min Neurons at the Edge: Pruned Autoencoder Implementation”! |
| Jun 19, 2023 | Second day of CVPR 2023!🍁 Happy to meet the Event-based community in Canada at the Vancouver Convetion Center!💥 Today I will present the paper “Pedro: An Event-based Dataset for Person Detection in Robotics”. Our dataset is freely available on GitHub! 🔓 |
| Dec 9, 2022 | What an unusual birthday🎂! Today I am spending my birthday in Saudi Arabia, visiting KAUST (King Abdullah University of Science and Technology) for the first time since my supervisor decided to move here! 🏝️ |
| Oct 13, 2022 | First time visiting Asia 🌏! I present today my paper “Aggresively prunable MAM-based Deep Neural Oracle for ECG acquisition by Compressed Sensing” 📝🫀 here in Taiwan! |
| Jul 11, 2022 | Hi from Brixen (BZ)⛰️! I will spend this week at the IEEE/DEI Summer Ph.D. School of Information Engineering “Silvano Pupolin” 🏫 following the Track no.1 about “Machine Learning, Sustainable Edge Computing”🌳📱! |
| May 23, 2022 | It’s ICRA time!🤖 Today I have presented here in Philadelphia my paper “Visual Navigation using Sparse Optical Flow and Time-To-Transit” 📷 written under the supervision of Prof. John Baillieul during my stay at Boston University!🎓 |
| Nov 1, 2021 | Today is the first day for me as a PhD student at Politecnico di Torino under the supervision of Professor Gianluca Setti! 🥳 My focus will be on Low-energy DNNs!🔋🤖 |
Publications
On the Universal Approximation Properties of Deep Neural Networks using MAM Neurons
IEEE TPAMI (Under review) - Preprint available here.
2024
Memory in Motion: Exploring Leaky Integration of Time Surfaces for Event-Based Eye-Tracking
BioCAS - The link to this publication will be available soon.
2024
A Multiply-And-Max/min Neuron Paradigm for Aggressively Prunable Deep Neural Networks
IEEE TNNLS (Under minor review) - Preprint available here.
2024
Optimizing Vision Transformers: Leveraging Max and Min Operations for Efficient Pruning
AICAS - You can read this publication here!
2024
Pedro: an Event-based Dataset for Person Detection in Robotics
CVPR - You can read this publication here!
2023
Multiply-And-Max/min Neurons at the Edge: Pruned Autoencoder Implementation
MWSCAS - You can read this publication here!
2023
Visual Navigation Using Sparse Optical Flow and Time-To-Transit
ICRA - You can read this publication here!
2022
Aggresively prunable MAM²-based Deep Neural Oracle for ECG acquisition by Compressed Sensing
BioCAS - You can read this publication here!
2022