Program for stream Large scale optimization: methods and algorithms

MB-03: First-order methods in modern optimization (Part I)
Stream: Large scale optimization: methods and algorithms
Room: B100/4011
Chair(s): Simone Rebegoldi, Andrea Sebastiani

• Learning firmly nonexpansive operators
  Jonathan Chirinos Rodriguez, Emanuele Naldi, Kristian Bredies
• Line search based stochastic gradient methods for learning applications
  Federica Porta
• Convergence rates of regularized quasi-Newton methods without strong convexity
  Shida Wang, Jalal Fadili, Peter Ochs
• LoCoDL: Communication-Efficient Distributed Optimization with Local Training and Compression
  Laurent Condat, Arto Maranjyan, Peter Richtarik

MC-03: First-order methods in modern optimization (Part II)
Stream: Large scale optimization: methods and algorithms
Room: B100/4011
Chair(s): Simone Rebegoldi, Andrea Sebastiani

• Optimization Techniques for Learning Multi-Index Models
  Hippolyte Labarrière, Shuo Huang, Ernesto De Vito, Lorenzo Rosasco, Tomaso Poggio
• Neural Blind Deconvolution for Poisson Data
  Alessandro Benfenati, Ambra Catozzi, Valeria Ruggiero
• Adaptively Inexact Bilevel Learning via Primal-Dual Differentiation
  Mohammad Sadegh Salehi
• Alternate Through the Epochs Stochastic Gradient for Multi-Task Neural Networks
  Stefania Bellavia, Francesco Della Santa, Alessandra Papini

TB-03: Theoretical and algorithmic advances in large scale nonlinear optimization and applications Part 1
Stream: Large scale optimization: methods and algorithms
Room: B100/4011
Chair(s): Stefania Bellavia, Benedetta Morini

• Fully stochastic trust-region methods with Barzilai-Borwein steplengths
  Benedetta Morini, Mahsa Yousefi, Stefania Bellavia
• prunAdag: an adaptive pruning-aware gradient method
  Giovanni Seraghiti, Margherita Porcelli, Philippe L. Toint
• An acceleration strategy for gradient methods in convex quadratic programming
  Gerardo Toraldo, Serena Crisci, Anna De Magistris, Valentina De Simone
• Corrective Frank-Wolfe: Unifying and Extending Correction Steps
  Jannis Halbey, Seta Rakotomandimby, Mathieu Besançon, Sebastian Pokutta

TB-10: First order methods: new perspectives for machine learning
Stream: Large scale optimization: methods and algorithms
Room: B100/8011
Chair(s): Cesare Molinari, Silvia Villa, Lorenzo Rosasco

• Convergence Analysis of Nonlinear Parabolic PDE Models with Neural Network Terms Trained with Gradient Descent
  Konstantin Riedl, Justin Sirignano, Konstantinos Spiliopoulos
• Randomized trust-region method for non-convex mimimization
  Radu-Alexandru Dragomir
• Perspectives on the analysis and design of optimization algorithms: Lyapunov analyses and counter-examples
  Adrien Taylor
• Accelerated Gradient Methods via Inertial Systems with Hessian-driven Damping
  Juan Peypouquet

TC-03: Theoretical and algorithmic advances in large scale nonlinear optimization and applications Part 2
Stream: Large scale optimization: methods and algorithms
Room: B100/4011
Chair(s): Stefania Bellavia, Benedetta Morini

• Advanced Techniques for Portfolio Optimization Under Uncertainty
  Valentina De Simone
• Inexact derivative-free methods for constrained bilevel optimization with applications to machine learning
  Marco Viola, Matteo Pernini, Gabriele Sanguin, Francesco Rinaldi
• Three Alternating Projection Methods for Matrix Completion
  Mattia Silei, Stefania Bellavia, Simone Rebegoldi
• Bi- and Multi-Level Optimization Strategies for Sparse and Interpretable Learning in NMF
  Laura Selicato

WB-03: Recent Advances in Line-Search Based Optimization
Stream: Large scale optimization: methods and algorithms
Room: B100/4011
Chair(s): Matteo Lapucci

• Line Search Methods are Sharpness-Aware and Operate at the Edge of Stability
  Leonardo Galli
• A Gradient Method with Momentum for Riemannian Manifolds
  Diego Scuppa, Filippo Leggio, Marco Sciandrone
• A Variable Dimension Sketching Strategy for Nonlinear Least-Squares
  Greta Malaspina, Stefania Bellavia, Benedetta Morini
• Stochastic line-search-based optimization for training overparameterized models: convergence conditions and effective approaches to leverage momentum
  Davide Pucci, Matteo Lapucci

WB-08: Theoretical advances in nonconvex optimization
Stream: Large scale optimization: methods and algorithms
Room: B100/7007
Chair(s): Annette Dumas, Clément Royer

• Continuized Nesterov Acceleration to improve convergence speed in non convex optimization
  Julien Hermant, Jean-François Aujol, Charles Dossal, Aude Rondepierre
• Complexity of Newton-type methods with quadratic regularization for nonlinear least squares
  Iskander LEGHERABA, Clément Royer
• Cubic regularized Newton methods with stochastic Hessian evaluations and momentum-based variance reduction
  Yiming Yang
• Algorithms for nonconvex optimization on measure spaces.
  Annette Dumas, Clément Royer

WC-03: Acceleration Methods in Optimization
Stream: Large scale optimization: methods and algorithms
Room: B100/4011
Chair(s): Vuong Phan, yingxin zhou

• Anderson acceleration with adaptive relaxation for convergent fixed-point iterations
  Nicolas Lepage-Saucier
• Anderson Acceleration for Primal-Dual Hybrid Gradient
  yingxin zhou, Stefano Cipolla, Vuong Phan
• Accelerating Convergence of MPGP Algorithm
  Jakub Kruzik, David Horak