Biography

Cesar Federico Caiafa is an Adjunct Professor in the Engineering Department at the University of Buenos Aires (FIUBA). He is also an independent researcher at IAR (Instituto Argentino de Radioastronomía) and CONICET (National Council for Scientific and Technical Research). After graduating in electronics engineering in 1996, he obtained a doctorate in engineering from the Faculty of Engineering at the University of Buenos Aires in December 2007. His work focuses on tensor factorizations, parsimonious representations and their applications in astronomy, biomedicine and neuroscience.

Collaboration with LPENSL

As part of an IEA (International Emerging Action) project with Julien Tachella from the SiSyPh team, Cesar Federico Caiafa will be working on new learning methods for solving the nonlinear inverse problem in microwave tomography (MWT), without the need for ground-truth training data. Three axes will be explored:

• New “unrolled” architectures for microwave tomography;
• Self-supervised learning methods adapted to MWT;
• Evaluation of methods on simulated ankle imaging data and integration into the deepinv library.

Anticipated contributions include:

• scientific collaborations with Nelly Pustelnik (unrolled networks) and Pierre Borgnat (signal processing for neuroscience);
• public conference on tensorial decomposition and parsimonious models;
• specialized seminar at MLSP (ENSL & Inria) on learning from incomplete data;
• doctoral courses on tensorial methods in machine learning and signal processing.

Cesar Federico Caiafa's visit aims to strengthen links between the French and Argentinean computational imaging communities, and to initiate long-term collaborations, such as the setting up of an international research network project (IRN-CNRS).

Major publications

• Bai B, Huang W, Li T, Wang A, Gao J, Caiafa CF, Zhao Q, “Diffusion Models Demand Contrastive Guidance for
  Adversarial Purification to Advance”, ICML 2024 (Forty-First International Conference on Machine Learning),
  Vienna, Austria.
• Zeng J, Li C, Caiafa CF, Zhao Q, "Alternating Local Enumeration (TnALE): Solving Tensor Network Structure
  Search with Fewer Evaluations”, C Li, ICML 2023 (Fortieth International Conference on Machine Learning),
  Honolulu, USA.
• Zhang J, Sun Z, Duan F, Shi L, Zhang Y, Solé-Casals J, Caiafa CF, “Cerebral cortex layer segmentation using
  diffusion magnetic resonance imaging in vivo with applications to laminar connections and working memory
  analysis”, Human Brain Mapping, 43(17), 2022.
• Caiafa, CF, Wang Z, Solé-Casals J, Zhao Q, “Learning from Incomplete Features by Simultaneous Training of
  Neural Networks and Sparse Coding” LLID Workshop at CVPR 2021 (Conference on Computer Vision and
  Pattern Recognition), New York, USA.
• Aminmansour F, Patterson A, Le L, Peng Y, Mitchell D, Pestilli F, Caiafa CF, Greiner R, White M, “Learning
  Macroscopic Brain Connectomes via Group-Sparse Factorization”. NeurIPS 2019 (Annual Conference on Neural
  Information Processing Systems), Vancouver, Canada.
• Avesani P, Caiafa CF, McPherson B, Hayashi S, Henschel R, Garyfallidis E, Patterson A, Sporns O, Saykin A,
  Pestilli F, “Derivatives as data: An open repository of repeated-measures structural connectomes and white matter
  tracts anatomy”, Nature Scientific Data 6(69), 2019.
• “Unified representation of tractography and diffusion-weighted MRI data using sparse multidimensional arrays”,
  Caiafa CF, Sporns O, Saykin AJ, Pestilli F, NeurIPS 2017 (Annual Conference on Neural Information Processing
  Systems), Long Beach, USA.
• Cichocki A, Mandic D, De Lathauwer L, Zhou G, Zhao Q; Caiafa CF, Phan HP, "Tensor decompositions for signal
  processing applications: From two-way to multiway component analysis" IEEE signal processing magazine 32.2
  (2015): 145-163.
• Caiafa, CF, and Cichocki A, "Computing sparse representations of multidimensional signals using Kronecker
  bases." Neural computation 25.1 (2013): 186-220.
  Caiafa CF, Cichocki A, “Generalizing the Column-Row Matrix Decomposition to Multi-way Arrays”, Linear Algebra and its Applications, 433 (2010): 557–573.