Hi! PARIS Meet Up! on Frugal AI
The second Hi! PARIS Meet Up! event will be dedicated to “Frugal AI”. The event will be hosted at Schneider Electric (RER A Rueil-Malmaison).
A great opportunity to connect to the latest research and to network with business leaders and academics.
Come and join us !
2ND MEET UP! ON FRUGAL AI
Thursday 15 June, 6:00-9:00 PM (Paris Time)
Auditorium at Schneider Electric SE. 35 rue Joseph Monier, 92 500 Rueil Malmaison (RER A)
This event is open to all who are interested in the topic and willing to take part in the discussion (business leaders, R&D / R&T / R&I Teams, Data scientists, Prospectives team, policy makers…).
How to access the venue?
The parking space request at Schneider Electric is included in the registration form. An email will be sent to the requesters with the details.
A public parking “Parking Indigo Rueil-Malmaison 2 Gares” is available at Residence Helium, 7 rue Amédée Bollée. Access map
Registration is mandatory for this in-person event.
Frugal AI – Less data, less energy, same performance?
As the performance of AI models increased in the recent decades, so did the size of the data, as well as the time and computational resources dedicated to training these models. Frugal AI has emerged as a series of strategies that aim to reduce both energy consumption and size of the data necessary for training, while not compromising on the quality and robustness of algorithms. One aspect of frugality has been made possible by the use of pre-trained algorithms, on which only fine-tuning is necessary, instead of retraining the whole model. Another one consists in improving the design of optimization algorithms and artificial intelligence models so that they can use data and computational resources more efficiently.
Join us for presentations and panel discussion of Hi! PARIS researchers in the domain and let’s discuss of Frugal AI and beyond!
Take the opportunity to meet young researchers from Hi! PARIS network who will present their PhD thesis through posters.
Program
Welcome to the venue from 6:00 PM.
The event will officially start at 6:15 PM with a poster session: our young researchers will present their doctoral thesis in a very accessible manner ; the CNIL will also have a corner to nurture our thinking around the topic of frugal AI.
From 7:15 PM, we will gather for a panel discussion with Hi! PARIS researchers experts in this domain and a Q&A session with the audience.
– Florence d’Alché (Telecom Paris – IP Paris)
– Philippe Ciblat (Telecom Paris – IP Paris)
– Michalis Vazirgiannis (École polytechnique – IP Paris)
From 8:15-9:00 PM, we will have the opportunity to extend the discussion with a networking cocktail on the terrace of the venue.
Registration
Registration is free and mandatory.
The speakers
The panel of the round table is composed of experts within these fields from Hi! PARIS institutions. They will give short presentations, as well as participate to a general round table discussion around the topics of Frugal AI!
Florence D’Alché-Buc, Télécom Paris – IP Paris
Florence d’Alché-Buc has been since 2014 a full professor at Télécom Paris, an IMT grande école. Since January 2019, she holds the Data Science and Artificial Intelligence for Digitalized Industry and Services Research and Teaching Chair. Since September 2021, she has headed the Images, Data, Signals department. Her research interests are machine learning, network inference, structured prediction and dynamical system modeling.
Philippe Ciblat, Télécom Paris – IP Paris
Philippe Ciblat has been a professor at Télécom Paris since 2001. From 2009 to 2020, he was the head of Digital Communications Team. His research areas include statistical signal processing, signal processing for digital communications, resource allocation, distributed optimization, signal over graphs, and machine learning.
Michalis Vazirgiannis, École polytechnique – IP Paris
Michalis Vazirgiannis is a Professor at LIX, Ecole Polytechnique in France and leads the Data Science and Mining group (DaSciM). His current research interests are on machine/deep learning and combinatorial methods for Graph analysis (including community detection, graph clustering, node embeddings and influence maximization), Text mining (including Graph of Words, word embeddings with applications to web advertising and marketing, event detection and summarization).
The poster presenters
The event will start with a poster session giving the floor to five of our young promising researchers from Hi! PARIS institutions, in the area of Artificial Intelligence and Data Science.
- Maxime BEDOIN, PhD Student at Télécom SudParis, IP Paris “Graph Neural Networks for improved EEG analysis in the context of Alzheimer’s disease”
- Fatma KIRAZ, PhD Student at Télécom Paris, IP Paris “Toward Robust Analog Equilibrium Propagation by Investigating Power Calculation Dynamics”
- Imad Eddine MAROUF, PhD Student at Télécom Paris, IP Paris “A Parameter-Efficient Transfer Learning Framework with Mini Adapters for Vision Transformers”
- Marie REINBIGLER, PhD Student at Télécom SudParis, IP Paris “Frugal Deep Learning for Multi-Resolution Image Analysis”