- Amphi. Jean-Jaurès, 29 rue d'Ulm, 21~22 September 2017, 8:55~17:20
Program Schedule (Tentative)
All invited talks will be 25 minutes long. Student contributed talks will be 10 minutes.
- Talks 21 AM
- Talks 21 PM
- Talks 22 AM
- Talks 22 PM
- 8:55 – 9:20 opening & RIKEN AIP presentation
- 9:20 – 9:45 Ichiro Takeuchi, NagoyaTech
Fitting and Testing Sparse High-Order Interaction Models (more…)
- 9:45 – 10:10 Joseph Salmon, Telecom
From safe screening rules to working sets for faster Lasso-type solvers
(more…) - 10:10 – 10:40 ——— coffee ———
- 10:40 – 11:05 Jean-Philippe Vert, ENS / Mines / Curie
Learning on the symmetric group (more…)
- 11:05 – 11:30 Koji Tsuda, U. of Tokyo / RIKEN AIP
Automatic design of functional molecules and materials (more…)
- 11:30 – 11:55 Chloé Azencott, Curie / Mines / INSERM
Network-guided high-dimensional feature selection in precision medicine (more…)
- 11:55 – 13:50 ——— Lunch – buffet served onsite
- 13:50 – 14:00 student talk 1: Mathieu Carrière
Sliced Wasserstein Kernels for Persistence Diagrams (more…)
- 14:00 – 14:10 student talk 2: Arthur Pajot
Deep Learning for Physical Processes: An Application to Sea Surface Temperature Forecasting (more…)
- 14:10 – 14:20 student talk 3: Adil Salim
Convergence of a constant step stochastic proximal gradient algorithm with generalization to random monotone operators (more…)
- 14:20 – 14:45 Junya Honda, U. of Tokyo / RIKEN AIP
Bandit Problems for Pairwise Feedback Models (more…)
- 14:45 – 15:10 Vianney Perchet, ENS Saclay
Fast Rates for Bandit Optimization with Upper-Confidence Frank-Wolfe (more…)
- 15:10 – 15:40 ——— coffee
- 15:40 – 16:05 Shohei Shimizu, Shiga U. / RIKEN AIP
Causal discovery and prediction mechanisms (more…)
- 16:05 – 16:30 Judith Rousseau, Oxford U. / U. Paris Dauphine
Using asymptotics to understand ABC (more…)
- 16:30 – 16:55 Kohei Hatano, Kyushu U. / RIKEN AIP
Boosting the kernelized shapelets: Theory and algorithms for local features (more…)
- 16:55 – 17:20 Gabriel Peyré, ENS Ulm
Optimal Transport and Deep Generative Models (more…)
- 8:55 – 9:20 Arnak Dalalyan, CREST, U. Paris-Saclay
User-friendly error bounds for sampling from a strongly log-concave density (more…)
- 9:20 – 9:45 Akiko Takeda, ISM / RIKEN AIP
Proximal DC Algorithm for Sparse Optimization (more…)
- 9:45 – 10:10 Francis Bach, ENS/Inria Paris
Optimal algorithms for smooth and strongly convex distributed optimization in networks (more…)
- 10:10 – 10:40 ——— coffee ———
- 10:40 – 11:05 Takanori Maehara, RIKEN AIP
Stochastic Packing Integer Programming with a Few Queries (more…)
- 11:05 – 11:30 Irène Waldspurger, U. Paris Dauphine
Phase retrieval with the alternating projections method (more…)
- 11:30 – 11:55 Taiji Suzuki, U. of Tokyo / RIKEN AIP
Generalization error bounds of deep learning by Bayesian and empirical risk minimization approaches from a kernel perspective (more…)
- 11:55 – 13:45 ——— Lunch – buffet served onsite
- 13:45 – 13:55 student talk 4: Anna Korba
A learning theory for ranking aggregation (more…)
- 13:55 – 14:20 Masashi Sugiyama, U. of Tokyo / RIKEN AIP
Classification from Weak Supervision (more…)
- 14:20 – 14:45 Robert Gower, Inria Paris
Stochastic Variance Reduced Methods Based on Sketching and Projecting (more…)
- 14:45 – 15:10 Naonori Ueda, NTT CS Labs / RIKEN AIP
Spatio-temporal collective data analysis for real-time and proactive navigation (more…)
- 15:10 – 15:40 ——— coffee ———
- 15:40 – 16:05 Ryota Tomioka, MSR Cambridge
AMPNet: Asynchronous Model-Parallel Training for Dynamic Neural Networks (more…)
- 16:05 – 16:30 Pierre Alquier, CREST, U. Paris Saclay
Concentration of variational approximations of posterior distributions (more…)
- 16:30 – 16:55 Mathieu Blondel, NTT CS Labs
A Regularized Framework for Sparse and Structured Neural Attention (more…)
- 16:55 – 17:20 Alexandre Gramfort, Inria Saclay
Faster independent component analysis by preconditioning with
Hessian approximations (more…)