News
Ten papers by ELLIS Genoa researchers @Neurips 2021
Members of Ellis Genoa had a number of accepted papers at NeurIPS 2021, the Thirty-fifth Conference on Neural Information Processing Systems.
Taking place virtually Dec 6th through Sat the 14th, NeurIPS is widely regarded as the most important international conference in AI. With ten accepted papers, the Ellis Genoa unit is among the top contributors among European research institutions and universities.
Nine papers by ELLIS Genoa researchers @Neurips 2020
Members of Ellis Genoa had a number of accepted papers at NeurIPS 2020, the Thirty-fourth Conference on Neural Information Processing Systems.
Taking place virtually Dec 6th through Sat the 12th, NeurIPS is widely regarded as the most important international conference in AI. With nine accepted papers, three of which presented as orals, the Ellis Genoa unit is the best represented contributor to the conference from italian research Institutions and universities. NeurIPS is a highly competitive conference, with 9454 submissions this year, of which about 20% were accepted, and only 105 receiving oral presentations.
IIT & UniGe Joint Machine Learning and Intelligent System Initiative
Genoa raises the bet on AI: a new agreement between the Italian Institute of Technology and the University of Genoa to become an increasingly leading player in Italy and Europe.
The news is still fresh that Genoa is one of the 30 nodes of Ellis, the European company of artificial intelligence and machine learning, including centers such as Cambridge, ETH Zurich, Oxford, and the Max Planck Institute. Now the Italian Institute of Technology and the University of Genoa are relaunching with a new agreement, a joint initiative on machine learning and artificial intelligence aimed at consolidating and expanding the activities already present with the aim of becoming the incubator of new research projects, training and technology transfer. The new agreement will facilitate exchanges and synergies between the two institutions. The researchers involved in the initiative are among the world leaders in AI with activities ranging from theoretical foundations and algorithms, to applications of machine learning in areas such as robotics, neuroscience and cognitive sciences, and healthcare and AI. The new agreement has the stated objective of making Genoa one of the AI hubs in Italy and Europe.
Papers
- ParK: Sound and Efficient Kernel Ridge Regression by Feature Space Partitions
Luigi Carratino · Stefano Vigogna · Daniele Calandriello · Lorenzo Rosasco - Distributed Zero-Order Optimization under Adversarial Noise
Aria Akhavan · Massimiliano Pontil · Alexandre Tsybakov - The Role of Global Labels in Few-Shot Classification and How to Infer Them.
Ruohan Wang · Massimiliano Pontil · Carlo Ciliberto - A Gang of Adversarial Bandits
Mark Herbster · Stephen None Pasteris · Fabio Vitale · Massimiliano Pontil - Concentration inequalities under sub-Gaussian and sub-exponential conditions
Andreas Maurer · Massimiliano Pontil - Implicit MLE: Backpropagating Through Discrete Exponential Family DistributionsMathias Niepert · Pasquale Minervini · Luca Franceschi
- Learning the optimal Tikhonov regularizer for inverse problems
Giovanni Alberti · Ernesto De Vito · Matti Lassas · Luca Ratti · Matteo Santacesaria - A New Theoretical Framework for Fast and Accurate Online Decision-Making
Nicolò Cesa-Bianchi · Tom Cesari · Yishay Mansour · Vianney Perchet - Beyond Bandit Feedback in Online Multiclass Classification
Dirk van der Hoeven · Federico Fusco · Nicolò Cesa-Bianchi
Papers
Oral
- Exact Recovery of Mangled Clusters with Same-Cluster Queries
Marco Bressan · Nicolò Cesa-Bianchi · Silvio Lattanzi · Andrea Paudice - Fair regression via plug-in estimator and recalibration with statistical guarantees
Evgenii Chzhen · Christophe Denis · Mohamed Hebiri · Luca Oneto · Massimiliano Pontil - Kernel Methods Through the Roof: Handling Billions of Points Efficiently
Giacomo Meanti · Luigi Carratino · Lorenzo Rosasco · Alessandro Rudi
Posters
- Exploiting Higher Order Smoothness in Derivative-free Optimization and Continuous Bandits
Arya Akhavan · Massimiliano Pontil · Alexandre Tsybakov - Fair regression with Wasserstein barycenters
Evgenii Chzhen · Christophe Denis · Mohamed Hebiri · Luca Oneto · Massimiliano Pontil - The Advantage of Conditional Meta-Learning for Biased Regularization and Fine Tuning
Giulia Denevi · Massimiliano Pontil · Carlo Ciliberto - Locally-Adaptive Nonparametric Online Learning
Ilja Kuzborskij · Nicolò Cesa-Bianchi - Estimating weighted areas under the ROC curve
Andreas Maurer · Massimiliano Pontil - Exploiting MMD and Sinkhorn Divergences for Fair and Transferable Representation Learning
Luca Oneto · Michele Donini · Giulia Luise · Carlo Ciliberto · Andreas Maurer · Massimiliano Pontil